rnn-chicago-transit-forecast-time-series-analysis.ipynb

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"## Recurrent Neural Networks - Chicago Transit Forecast.","metadata":{}},{"cell_type":"code","source":"import tensorflow as tf\nimport matplotlib.pyplot as plt\nfrom pathlib import Path\nimport pandas as pd\nimport numpy as np","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2024-07-11T17:24:37.094285Z","iopub.execute_input":"2024-07-11T17:24:37.094714Z","iopub.status.idle":"2024-07-11T17:24:58.252285Z","shell.execute_reply.started":"2024-07-11T17:24:37.094684Z","shell.execute_reply":"2024-07-11T17:24:58.250497Z"},"trusted":true},"execution_count":3,"outputs":[{"name":"stderr","text":"2024-07-11 17:24:40.373799: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n2024-07-11 17:24:40.374039: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n2024-07-11 17:24:40.654841: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n","output_type":"stream"}]},{"cell_type":"markdown","source":"# Loading the Dataset","metadata":{}},{"cell_type":"code","source":"path = Path(\"/kaggle/input/chicago-transit-authority-ridership-data/CTA_-_Ridership_-_Daily_Boarding_Totals.csv\")\n\ndf = pd.read_csv(path, parse_dates=[\"service_date\"])\ndf.columns = [\"date\", \"day_type\", \"bus\", \"rail\", \"total\"] # creating much shorter names\ndf = df.sort_values(\"date\").set_index(\"date\")\ndf = df.drop(\"total\", axis=1) # no need for the column as its just bus + rail\ndf = df.drop_duplicates() # remocing the duplicate months present in the dataset.","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:24:58.254497Z","iopub.execute_input":"2024-07-11T17:24:58.255319Z","iopub.status.idle":"2024-07-11T17:24:58.388336Z","shell.execute_reply.started":"2024-07-11T17:24:58.255281Z","shell.execute_reply":"2024-07-11T17:24:58.386420Z"},"trusted":true},"execution_count":4,"outputs":[]},{"cell_type":"code","source":"df.head()","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:24:58.390702Z","iopub.execute_input":"2024-07-11T17:24:58.391204Z","iopub.status.idle":"2024-07-11T17:24:58.413685Z","shell.execute_reply.started":"2024-07-11T17:24:58.391168Z","shell.execute_reply":"2024-07-11T17:24:58.411920Z"},"trusted":true},"execution_count":5,"outputs":[{"execution_count":5,"output_type":"execute_result","data":{"text/plain":"           day_type     bus    rail\ndate                               \n2001-01-01        U  297192  126455\n2001-01-02        W  780827  501952\n2001-01-03        W  824923  536432\n2001-01-04        W  870021  550011\n2001-01-05        W  890426  557917","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>day_type</th>\n      <th>bus</th>\n      <th>rail</th>\n    </tr>\n    <tr>\n      <th>date</th>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>2001-01-01</th>\n      <td>U</td>\n      <td>297192</td>\n      <td>126455</td>\n    </tr>\n    <tr>\n      <th>2001-01-02</th>\n      <td>W</td>\n      <td>780827</td>\n      <td>501952</td>\n    </tr>\n    <tr>\n      <th>2001-01-03</th>\n      <td>W</td>\n      <td>824923</td>\n      <td>536432</td>\n    </tr>\n    <tr>\n      <th>2001-01-04</th>\n      <td>W</td>\n      <td>870021</td>\n      <td>550011</td>\n    </tr>\n    <tr>\n      <th>2001-01-05</th>\n      <td>W</td>\n      <td>890426</td>\n      <td>557917</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"markdown","source":"## Light EDA","metadata":{}},{"cell_type":"code","source":"df[\"2019-03\":\"2019-05\"].plot(grid=True, marker=\".\", figsize=(8, 3.5))\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:24:58.418110Z","iopub.execute_input":"2024-07-11T17:24:58.418690Z","iopub.status.idle":"2024-07-11T17:24:59.108819Z","shell.execute_reply.started":"2024-07-11T17:24:58.418653Z","shell.execute_reply":"2024-07-11T17:24:59.107157Z"},"trusted":true},"execution_count":6,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 800x350 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"markdown","source":"- Data seems to be consistent across the whole time, we'll try to plot it together with the previous weeks data and show the difference - differencing","metadata":{}},{"cell_type":"code","source":"diff_7 = df[[\"bus\", \"rail\"]].diff(7)[\"2019-03\":\"2019-05\"]\n\nfig, axs = plt.subplots(2, 1, sharex=True, figsize=(8, 5))\ndf.plot(ax=axs[0],legend=False, marker=\".\") #original time series.\ndf.shift(7).plot(ax=axs[0], grid=True, legend=False, linestyle=\":\") # the lagged time series.\ndiff_7.plot(ax=axs[1], grid=True, marker=\".\") # the 7-day difference time series.\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:24:59.110914Z","iopub.execute_input":"2024-07-11T17:24:59.111449Z","iopub.status.idle":"2024-07-11T17:25:00.698749Z","shell.execute_reply.started":"2024-07-11T17:24:59.111408Z","shell.execute_reply":"2024-07-11T17:25:00.696744Z"},"trusted":true},"execution_count":7,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 800x500 with 2 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"markdown","source":"- We can see that the time series is autocorrelated.\n- Monday was a long holiday that's why the gap\n- Next lets measure the mean absolute error over the three month period ( March - May )","metadata":{}},{"cell_type":"markdown","source":"## Calculating the mean absolute error","metadata":{}},{"cell_type":"code","source":"diff_7.abs().mean()","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:00.701053Z","iopub.execute_input":"2024-07-11T17:25:00.701592Z","iopub.status.idle":"2024-07-11T17:25:00.716062Z","shell.execute_reply.started":"2024-07-11T17:25:00.701546Z","shell.execute_reply":"2024-07-11T17:25:00.714269Z"},"trusted":true},"execution_count":8,"outputs":[{"execution_count":8,"output_type":"execute_result","data":{"text/plain":"bus     43915.608696\nrail    42143.271739\ndtype: float64"},"metadata":{}}]},{"cell_type":"markdown","source":"## Calculating the mean absolute percentage error.","metadata":{}},{"cell_type":"code","source":"# we will put the the forecasts in perspective by dividing them by the target values.\ntargets = df[[\"bus\", \"rail\"]][\"2019-03\":\"2019-05\"]\n(diff_7 / targets).abs().mean()","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:00.717975Z","iopub.execute_input":"2024-07-11T17:25:00.718423Z","iopub.status.idle":"2024-07-11T17:25:00.738433Z","shell.execute_reply.started":"2024-07-11T17:25:00.718390Z","shell.execute_reply":"2024-07-11T17:25:00.736504Z"},"trusted":true},"execution_count":9,"outputs":[{"execution_count":9,"output_type":"execute_result","data":{"text/plain":"bus     0.082938\nrail    0.089948\ndtype: float64"},"metadata":{}}]},{"cell_type":"markdown","source":"## Checking for Yearly seasonality","metadata":{}},{"cell_type":"code","source":"# to reduce the risk of data snooping we'll check only the earlier data and ignore the most recent.\n\n# Only selecting the numeric columns\nnumeric_cols = df.select_dtypes(include=['number']).columns\ndf_numeric = df[numeric_cols]\n\nperiod = slice(\"2001\", \"2019\")\ndf_monthly = df_numeric.resample('M').mean()  # compute the mean for each month\nrolling_average_12_months = df_monthly[period].rolling(window=12).mean()\n\nfig, ax = plt.subplots(figsize=(8, 4))\ndf_monthly[period].plot(ax=ax, marker=\".\")\nrolling_average_12_months.plot(ax=ax, grid=True, legend=False)\n\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:00.740764Z","iopub.execute_input":"2024-07-11T17:25:00.741636Z","iopub.status.idle":"2024-07-11T17:25:01.218339Z","shell.execute_reply.started":"2024-07-11T17:25:00.741576Z","shell.execute_reply":"2024-07-11T17:25:01.216276Z"},"trusted":true},"execution_count":10,"outputs":[{"name":"stderr","text":"/tmp/ipykernel_34/3324810861.py:8: FutureWarning: 'M' is deprecated and will be removed in a future version, please use 'ME' instead.\n  df_monthly = df_numeric.resample('M').mean()  # compute the mean for each month\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 800x400 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"# We have some yearly seasonlaity even though its noisier.\n# We'll try to check if we plot the 12-month difference.\n\ndf_monthly.diff(12)[period].plot(grid=True, marker=\".\", figsize=(8, 3))\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:01.221050Z","iopub.execute_input":"2024-07-11T17:25:01.221682Z","iopub.status.idle":"2024-07-11T17:25:02.032480Z","shell.execute_reply.started":"2024-07-11T17:25:01.221634Z","shell.execute_reply":"2024-07-11T17:25:02.030800Z"},"trusted":true},"execution_count":11,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 800x300 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"markdown","source":"- So far we've looked at the following concepts in time series analysis (seasonality, trends, differencing and moving averages)\n- The ARMA Family Model computes it's forecasts using simple weighted sum of lagged values and corrects these forecasts by adding moving average - the moving average component is computed using a weighted sum of the last few forecast errors.","metadata":{}},{"cell_type":"markdown","source":"# Fitting the SARIMA model to the rail time series.","metadata":{}},{"cell_type":"code","source":"# SARIMA is a member of the ARMA but seasonal, it models the time series in the same way as ARIMA, but it additionally models,\n# a seasonal component for a given frequency (e.g weekly)\n# We'll use it to make the 'tommorrow's; forecast.\n\nfrom statsmodels.tsa.arima.model import ARIMA\n\norigin, today = \"2019-01-01\", \"2019-05-31\"\nrail_series = df.loc[origin:today][\"rail\"].asfreq(\"D\")\nmodel = ARIMA(rail_series, \n              order=(1, 0, 0), \n              seasonal_order=(0, 1, 1, 7))\nmodel = model.fit()\ny_pred = model.forecast()","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:02.038278Z","iopub.execute_input":"2024-07-11T17:25:02.038759Z","iopub.status.idle":"2024-07-11T17:25:03.503729Z","shell.execute_reply.started":"2024-07-11T17:25:02.038714Z","shell.execute_reply":"2024-07-11T17:25:03.501436Z"},"trusted":true},"execution_count":12,"outputs":[]},{"cell_type":"code","source":"y_pred","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:03.507332Z","iopub.execute_input":"2024-07-11T17:25:03.509706Z","iopub.status.idle":"2024-07-11T17:25:03.527572Z","shell.execute_reply.started":"2024-07-11T17:25:03.509609Z","shell.execute_reply":"2024-07-11T17:25:03.525682Z"},"trusted":true},"execution_count":13,"outputs":[{"execution_count":13,"output_type":"execute_result","data":{"text/plain":"2019-06-01    427758.62641\nFreq: D, dtype: float64"},"metadata":{}}]},{"cell_type":"markdown","source":"- The actual value was 379,044 off by 12.9% which is slighly worse than the differencing at 12.6%\n- We'll try to run the same code in a loop to make forecasts fpr every day in March, April and May and compute the MAE over that period.","metadata":{}},{"cell_type":"code","source":"origin, start_date, end_date = \"2019-01-01\", \"2019-03-01\", \"2019-05-31\"\ntime_period = pd.date_range(start_date, end_date)\nrail_series = df.loc[origin:end_date][\"rail\"].asfreq(\"D\")\ny_preds = []\nfor today in time_period.shift(-1):\n    model = ARIMA(rail_series[origin:today], # train on the data upto today\n                  order=(1, 0, 0),\n                  seasonal_order=(0, 1, 1, 7))\n    model = model.fit() #NOTE: we retrain the model everyday.\n    y_pred = model.forecast().iloc[0]\n    y_preds.append(y_pred)\ny_preds = pd.Series(y_preds, index=time_period)\nmae = (y_preds - rail_series[time_period]).abs().mean()","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:03.531189Z","iopub.execute_input":"2024-07-11T17:25:03.533128Z","iopub.status.idle":"2024-07-11T17:25:21.781156Z","shell.execute_reply.started":"2024-07-11T17:25:03.533035Z","shell.execute_reply":"2024-07-11T17:25:21.779074Z"},"trusted":true},"execution_count":14,"outputs":[]},{"cell_type":"code","source":"mae","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:21.784183Z","iopub.execute_input":"2024-07-11T17:25:21.785650Z","iopub.status.idle":"2024-07-11T17:25:21.800389Z","shell.execute_reply.started":"2024-07-11T17:25:21.785572Z","shell.execute_reply":"2024-07-11T17:25:21.798471Z"},"trusted":true},"execution_count":15,"outputs":[{"execution_count":15,"output_type":"execute_result","data":{"text/plain":"32040.720089453378"},"metadata":{}}]},{"cell_type":"markdown","source":"- This is much better that all the other above, next we'll train ML models","metadata":{}},{"cell_type":"markdown","source":"# Preparing the Data for Machine Learning Models.","metadata":{}},{"cell_type":"code","source":"# Goal is to forecast tomorrows ridership based on the ridership of the past 8 weeks of data (56 days)\n# Inputs will be sequences each containong 56 values from time step t-55 to t.\n# For each input sequence, the model will output a single value: the forecast for time step t+1.\n# The training data we will use every 56-day window from the past as training data,target will be the value following it.\n# We will create our dataset using the window() method, which is more complex than the alternative, tf.keras.utils.timeseries_ data set_from_array()","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:21.804090Z","iopub.execute_input":"2024-07-11T17:25:21.805657Z","iopub.status.idle":"2024-07-11T17:25:21.852653Z","shell.execute_reply.started":"2024-07-11T17:25:21.805579Z","shell.execute_reply":"2024-07-11T17:25:21.850065Z"},"trusted":true},"execution_count":16,"outputs":[]},{"cell_type":"code","source":"dataset = tf.data.Dataset.range(6).window(4, shift=1, drop_remainder=True)\ndataset = dataset.flat_map(lambda window_dataset: window_dataset.batch(4))\nfor window_tensor in dataset:\n    print(f\"{window_tensor}\")","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:21.858645Z","iopub.execute_input":"2024-07-11T17:25:21.860860Z","iopub.status.idle":"2024-07-11T17:25:22.032111Z","shell.execute_reply.started":"2024-07-11T17:25:21.860774Z","shell.execute_reply":"2024-07-11T17:25:22.030055Z"},"trusted":true},"execution_count":17,"outputs":[{"name":"stdout","text":"[0 1 2 3]\n[1 2 3 4]\n[2 3 4 5]\n","output_type":"stream"}]},{"cell_type":"code","source":"# creating a helper function to make it easier to extract windows from a dataset.\n\ndef to_windows(dataset, length):\n    dataset = dataset.window(length, shift=1, drop_remainder=True)\n    return dataset.flat_map(lambda window_ds:window_ds.batch(length))","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:22.034460Z","iopub.execute_input":"2024-07-11T17:25:22.035050Z","iopub.status.idle":"2024-07-11T17:25:22.043649Z","shell.execute_reply.started":"2024-07-11T17:25:22.035001Z","shell.execute_reply":"2024-07-11T17:25:22.041531Z"},"trusted":true},"execution_count":18,"outputs":[]},{"cell_type":"code","source":"# Next step is to split each window into inputs and targets, using the map method. we can also group the resulting windows into batches of size 2:\n\ndataset = to_windows(tf.data.Dataset.range(6), 4) # 3 inputs + 1 target = 4\ndataset = dataset.map(lambda window: (window[:-1], window[-1]))\nlist(dataset.batch(2))","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:22.045909Z","iopub.execute_input":"2024-07-11T17:25:22.046521Z","iopub.status.idle":"2024-07-11T17:25:22.167713Z","shell.execute_reply.started":"2024-07-11T17:25:22.046443Z","shell.execute_reply":"2024-07-11T17:25:22.165817Z"},"trusted":true},"execution_count":19,"outputs":[{"execution_count":19,"output_type":"execute_result","data":{"text/plain":"[(<tf.Tensor: shape=(2, 3), dtype=int64, numpy=\n  array([[0, 1, 2],\n         [1, 2, 3]])>,\n  <tf.Tensor: shape=(2,), dtype=int64, numpy=array([3, 4])>),\n (<tf.Tensor: shape=(1, 3), dtype=int64, numpy=array([[2, 3, 4]])>,\n  <tf.Tensor: shape=(1,), dtype=int64, numpy=array([5])>)]"},"metadata":{}}]},{"cell_type":"code","source":"# we need to split our data into a training\n# period, a validation period, and a test period. We will focus on the rail\n# ridership for now. We will also scale it down by a factor of one million, to\n# ensure the values are near the 0–1 range; this plays nicely with the default\n# weight initialization and learning rate:","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:22.169457Z","iopub.execute_input":"2024-07-11T17:25:22.170087Z","iopub.status.idle":"2024-07-11T17:25:22.176818Z","shell.execute_reply.started":"2024-07-11T17:25:22.170031Z","shell.execute_reply":"2024-07-11T17:25:22.175020Z"},"trusted":true},"execution_count":20,"outputs":[]},{"cell_type":"code","source":"rail_train = df[\"rail\"][\"2016-01\":\"2018-12\"] / 1e6\nrail_valid = df[\"rail\"][\"2019-01\":\"2019-05\"] / 1e6\nrail_test = df[\"rail\"][\"2019-06\":] / 1e6","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:22.178905Z","iopub.execute_input":"2024-07-11T17:25:22.179607Z","iopub.status.idle":"2024-07-11T17:25:22.196309Z","shell.execute_reply.started":"2024-07-11T17:25:22.179547Z","shell.execute_reply":"2024-07-11T17:25:22.194362Z"},"trusted":true},"execution_count":21,"outputs":[]},{"cell_type":"markdown","source":"## Creating the datasets for training and validation","metadata":{}},{"cell_type":"code","source":"seq_length = 56\n\ntrain_ds = tf.keras.utils.timeseries_dataset_from_array(\n    rail_train.to_numpy(),\n    targets=rail_train[seq_length:],\n    sequence_length=seq_length,\n    batch_size=32,\n    shuffle=True,\n    seed=42\n)\n\nvalid_ds = tf.keras.utils.timeseries_dataset_from_array(\n    rail_valid.to_numpy(),\n    targets = rail_valid[seq_length:],\n    sequence_length = seq_length,\n    batch_size=32\n)","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:22.197916Z","iopub.execute_input":"2024-07-11T17:25:22.198450Z","iopub.status.idle":"2024-07-11T17:25:22.488251Z","shell.execute_reply.started":"2024-07-11T17:25:22.198407Z","shell.execute_reply":"2024-07-11T17:25:22.486293Z"},"trusted":true},"execution_count":22,"outputs":[]},{"cell_type":"markdown","source":"# Forecasting using a Linear Model","metadata":{}},{"cell_type":"code","source":"# We will use the Huber loss which usually works better than minimizing the MAE directly, we'll also use early stopping.\n\ntf.random.set_seed(42)\nmodel = tf.keras.Sequential([\n    tf.keras.layers.Dense(1, input_shape=[seq_length])\n])\n\nearly_stopping_cb = tf.keras.callbacks.EarlyStopping(\n    monitor=\"val_mae\", patience=50, restore_best_weights=True\n)\n\nopt = tf.keras.optimizers.SGD(learning_rate=0.02, momentum=0.9)\n\nmodel.compile(loss=tf.keras.losses.Huber(), optimizer=opt, metrics=[\"mae\"])\n\nhistory = model.fit(train_ds, validation_data=valid_ds, epochs=500, callbacks=[early_stopping_cb])","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:25:22.490193Z","iopub.execute_input":"2024-07-11T17:25:22.490656Z","iopub.status.idle":"2024-07-11T17:26:29.546367Z","shell.execute_reply.started":"2024-07-11T17:25:22.490618Z","shell.execute_reply":"2024-07-11T17:26:29.544422Z"},"trusted":true},"execution_count":23,"outputs":[{"name":"stdout","text":"Epoch 1/500\n","output_type":"stream"},{"name":"stderr","text":"/opt/conda/lib/python3.10/site-packages/keras/src/layers/core/dense.py:87: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n","output_type":"stream"},{"name":"stdout","text":"\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 12ms/step - loss: 0.1525 - mae: 0.4492 - val_loss: 0.0104 - val_mae: 0.1091\nEpoch 2/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0149 - mae: 0.1356 - val_loss: 0.0068 - val_mae: 0.0887\nEpoch 3/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0098 - mae: 0.1037 - val_loss: 0.0055 - val_mae: 0.0757\nEpoch 4/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0078 - mae: 0.0913 - val_loss: 0.0049 - val_mae: 0.0688\nEpoch 5/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0069 - mae: 0.0838 - val_loss: 0.0043 - val_mae: 0.0632\nEpoch 6/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0077 - mae: 0.0930 - val_loss: 0.0049 - val_mae: 0.0705\nEpoch 7/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0062 - mae: 0.0778 - val_loss: 0.0045 - val_mae: 0.0681\nEpoch 8/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0060 - mae: 0.0765 - val_loss: 0.0046 - val_mae: 0.0715\nEpoch 9/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0058 - mae: 0.0762 - val_loss: 0.0035 - val_mae: 0.0535\nEpoch 10/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0053 - mae: 0.0718 - val_loss: 0.0032 - val_mae: 0.0503\nEpoch 11/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0052 - mae: 0.0699 - val_loss: 0.0034 - val_mae: 0.0551\nEpoch 12/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0052 - mae: 0.0719 - val_loss: 0.0041 - val_mae: 0.0663\nEpoch 13/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0052 - mae: 0.0723 - val_loss: 0.0033 - val_mae: 0.0537\nEpoch 14/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0043 - mae: 0.0632 - val_loss: 0.0040 - val_mae: 0.0656\nEpoch 15/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0044 - mae: 0.0644 - val_loss: 0.0032 - val_mae: 0.0542\nEpoch 16/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0040 - mae: 0.0619 - val_loss: 0.0033 - val_mae: 0.0564\nEpoch 17/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0044 - mae: 0.0628 - val_loss: 0.0027 - val_mae: 0.0439\nEpoch 18/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0041 - mae: 0.0611 - val_loss: 0.0029 - val_mae: 0.0494\nEpoch 19/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0044 - mae: 0.0649 - val_loss: 0.0026 - val_mae: 0.0431\nEpoch 20/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0039 - mae: 0.0581 - val_loss: 0.0026 - val_mae: 0.0426\nEpoch 21/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0039 - mae: 0.0577 - val_loss: 0.0031 - val_mae: 0.0523\nEpoch 22/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0040 - mae: 0.0573 - val_loss: 0.0026 - val_mae: 0.0430\nEpoch 23/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0040 - mae: 0.0597 - val_loss: 0.0026 - val_mae: 0.0434\nEpoch 24/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0038 - mae: 0.0558 - val_loss: 0.0025 - val_mae: 0.0423\nEpoch 25/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0039 - mae: 0.0599 - val_loss: 0.0025 - val_mae: 0.0424\nEpoch 26/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0037 - mae: 0.0553 - val_loss: 0.0025 - val_mae: 0.0425\nEpoch 27/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0036 - mae: 0.0550 - val_loss: 0.0025 - val_mae: 0.0433\nEpoch 28/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0035 - mae: 0.0531 - val_loss: 0.0029 - val_mae: 0.0522\nEpoch 29/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0034 - mae: 0.0546 - val_loss: 0.0031 - val_mae: 0.0511\nEpoch 30/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0038 - mae: 0.0566 - val_loss: 0.0025 - val_mae: 0.0413\nEpoch 31/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0039 - mae: 0.0569 - val_loss: 0.0025 - val_mae: 0.0434\nEpoch 32/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0039 - mae: 0.0590 - val_loss: 0.0024 - val_mae: 0.0415\nEpoch 33/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0035 - mae: 0.0539 - val_loss: 0.0025 - val_mae: 0.0410\nEpoch 34/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0037 - mae: 0.0540 - val_loss: 0.0024 - val_mae: 0.0409\nEpoch 35/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0032 - mae: 0.0496 - val_loss: 0.0024 - val_mae: 0.0409\nEpoch 36/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0036 - mae: 0.0543 - val_loss: 0.0024 - val_mae: 0.0407\nEpoch 37/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0036 - mae: 0.0529 - val_loss: 0.0024 - val_mae: 0.0409\nEpoch 38/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0031 - mae: 0.0478 - val_loss: 0.0026 - val_mae: 0.0432\nEpoch 39/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0036 - mae: 0.0529 - val_loss: 0.0024 - val_mae: 0.0407\nEpoch 40/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0031 - mae: 0.0486 - val_loss: 0.0024 - val_mae: 0.0409\nEpoch 41/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0031 - mae: 0.0485 - val_loss: 0.0028 - val_mae: 0.0467\nEpoch 42/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0031 - mae: 0.0499 - val_loss: 0.0026 - val_mae: 0.0431\nEpoch 43/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0033 - mae: 0.0503 - val_loss: 0.0024 - val_mae: 0.0414\nEpoch 44/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0498 - val_loss: 0.0024 - val_mae: 0.0416\nEpoch 45/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0034 - mae: 0.0492 - val_loss: 0.0023 - val_mae: 0.0398\nEpoch 46/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0032 - mae: 0.0554 - val_loss: 0.0024 - val_mae: 0.0403\nEpoch 47/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0036 - mae: 0.0515 - val_loss: 0.0029 - val_mae: 0.0544\nEpoch 48/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0032 - mae: 0.0493 - val_loss: 0.0028 - val_mae: 0.0529\nEpoch 49/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0028 - mae: 0.0499 - val_loss: 0.0024 - val_mae: 0.0400\nEpoch 50/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0035 - mae: 0.0544 - val_loss: 0.0030 - val_mae: 0.0505\nEpoch 51/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0035 - mae: 0.0576 - val_loss: 0.0024 - val_mae: 0.0400\nEpoch 52/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0469 - val_loss: 0.0024 - val_mae: 0.0421\nEpoch 53/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0027 - mae: 0.0461 - val_loss: 0.0026 - val_mae: 0.0429\nEpoch 54/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0031 - mae: 0.0479 - val_loss: 0.0028 - val_mae: 0.0470\nEpoch 55/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0033 - mae: 0.0494 - val_loss: 0.0028 - val_mae: 0.0516\nEpoch 56/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0034 - mae: 0.0521 - val_loss: 0.0024 - val_mae: 0.0409\nEpoch 57/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 0.0030 - mae: 0.0475 - val_loss: 0.0023 - val_mae: 0.0399\nEpoch 58/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 8ms/step - loss: 0.0030 - mae: 0.0469 - val_loss: 0.0024 - val_mae: 0.0408\nEpoch 59/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 8ms/step - loss: 0.0032 - mae: 0.0482 - val_loss: 0.0028 - val_mae: 0.0464\nEpoch 60/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0038 - mae: 0.0545 - val_loss: 0.0025 - val_mae: 0.0418\nEpoch 61/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0033 - mae: 0.0482 - val_loss: 0.0023 - val_mae: 0.0397\nEpoch 62/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 8ms/step - loss: 0.0030 - mae: 0.0462 - val_loss: 0.0025 - val_mae: 0.0417\nEpoch 63/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0035 - mae: 0.0495 - val_loss: 0.0023 - val_mae: 0.0394\nEpoch 64/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0030 - mae: 0.0507 - val_loss: 0.0026 - val_mae: 0.0467\nEpoch 65/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0026 - mae: 0.0448 - val_loss: 0.0027 - val_mae: 0.0456\nEpoch 66/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0040 - mae: 0.0629 - val_loss: 0.0028 - val_mae: 0.0521\nEpoch 67/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0028 - mae: 0.0453 - val_loss: 0.0027 - val_mae: 0.0439\nEpoch 68/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0030 - mae: 0.0482 - val_loss: 0.0026 - val_mae: 0.0434\nEpoch 69/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0032 - mae: 0.0497 - val_loss: 0.0024 - val_mae: 0.0397\nEpoch 70/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0024 - mae: 0.0440 - val_loss: 0.0023 - val_mae: 0.0398\nEpoch 71/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0031 - mae: 0.0470 - val_loss: 0.0023 - val_mae: 0.0395\nEpoch 72/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0027 - mae: 0.0427 - val_loss: 0.0023 - val_mae: 0.0392\nEpoch 73/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0030 - mae: 0.0456 - val_loss: 0.0024 - val_mae: 0.0401\nEpoch 74/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 8ms/step - loss: 0.0034 - mae: 0.0486 - val_loss: 0.0025 - val_mae: 0.0415\nEpoch 75/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 8ms/step - loss: 0.0031 - mae: 0.0460 - val_loss: 0.0025 - val_mae: 0.0444\nEpoch 76/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0036 - mae: 0.0530 - val_loss: 0.0024 - val_mae: 0.0415\nEpoch 77/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0031 - mae: 0.0533 - val_loss: 0.0027 - val_mae: 0.0448\nEpoch 78/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0033 - mae: 0.0520 - val_loss: 0.0026 - val_mae: 0.0482\nEpoch 79/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0458 - val_loss: 0.0028 - val_mae: 0.0473\nEpoch 80/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0033 - mae: 0.0487 - val_loss: 0.0033 - val_mae: 0.0550\nEpoch 81/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0031 - mae: 0.0520 - val_loss: 0.0023 - val_mae: 0.0407\nEpoch 82/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0027 - mae: 0.0444 - val_loss: 0.0024 - val_mae: 0.0423\nEpoch 83/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0028 - mae: 0.0464 - val_loss: 0.0023 - val_mae: 0.0389\nEpoch 84/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0468 - val_loss: 0.0027 - val_mae: 0.0447\nEpoch 85/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0034 - mae: 0.0503 - val_loss: 0.0023 - val_mae: 0.0386\nEpoch 86/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0454 - val_loss: 0.0024 - val_mae: 0.0424\nEpoch 87/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0025 - mae: 0.0422 - val_loss: 0.0025 - val_mae: 0.0410\nEpoch 88/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0031 - mae: 0.0471 - val_loss: 0.0025 - val_mae: 0.0406\nEpoch 89/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 8ms/step - loss: 0.0028 - mae: 0.0448 - val_loss: 0.0027 - val_mae: 0.0436\nEpoch 90/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0039 - mae: 0.0561 - val_loss: 0.0024 - val_mae: 0.0404\nEpoch 91/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0492 - val_loss: 0.0026 - val_mae: 0.0468\nEpoch 92/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0456 - val_loss: 0.0025 - val_mae: 0.0408\nEpoch 93/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0028 - mae: 0.0448 - val_loss: 0.0025 - val_mae: 0.0457\nEpoch 94/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0032 - mae: 0.0496 - val_loss: 0.0024 - val_mae: 0.0398\nEpoch 95/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0028 - mae: 0.0447 - val_loss: 0.0023 - val_mae: 0.0387\nEpoch 96/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 8ms/step - loss: 0.0029 - mae: 0.0442 - val_loss: 0.0023 - val_mae: 0.0398\nEpoch 97/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 0.0028 - mae: 0.0440 - val_loss: 0.0024 - val_mae: 0.0397\nEpoch 98/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0460 - val_loss: 0.0024 - val_mae: 0.0400\nEpoch 99/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0034 - mae: 0.0482 - val_loss: 0.0023 - val_mae: 0.0387\nEpoch 100/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0447 - val_loss: 0.0023 - val_mae: 0.0389\nEpoch 101/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0459 - val_loss: 0.0023 - val_mae: 0.0388\nEpoch 102/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0028 - mae: 0.0455 - val_loss: 0.0024 - val_mae: 0.0402\nEpoch 103/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0028 - mae: 0.0447 - val_loss: 0.0023 - val_mae: 0.0391\nEpoch 104/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0460 - val_loss: 0.0024 - val_mae: 0.0426\nEpoch 105/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0451 - val_loss: 0.0026 - val_mae: 0.0478\nEpoch 106/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0030 - mae: 0.0479 - val_loss: 0.0023 - val_mae: 0.0390\nEpoch 107/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0028 - mae: 0.0446 - val_loss: 0.0024 - val_mae: 0.0414\nEpoch 108/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0031 - mae: 0.0468 - val_loss: 0.0023 - val_mae: 0.0389\nEpoch 109/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0029 - mae: 0.0452 - val_loss: 0.0023 - val_mae: 0.0388\nEpoch 110/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0028 - mae: 0.0433 - val_loss: 0.0025 - val_mae: 0.0421\nEpoch 111/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0469 - val_loss: 0.0023 - val_mae: 0.0407\nEpoch 112/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0028 - mae: 0.0456 - val_loss: 0.0023 - val_mae: 0.0391\nEpoch 113/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0027 - mae: 0.0450 - val_loss: 0.0023 - val_mae: 0.0390\nEpoch 114/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0027 - mae: 0.0445 - val_loss: 0.0024 - val_mae: 0.0395\nEpoch 115/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0031 - mae: 0.0452 - val_loss: 0.0025 - val_mae: 0.0454\nEpoch 116/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0025 - mae: 0.0425 - val_loss: 0.0026 - val_mae: 0.0418\nEpoch 117/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0028 - mae: 0.0458 - val_loss: 0.0023 - val_mae: 0.0385\nEpoch 118/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0028 - mae: 0.0436 - val_loss: 0.0030 - val_mae: 0.0515\nEpoch 119/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0026 - mae: 0.0450 - val_loss: 0.0023 - val_mae: 0.0392\nEpoch 120/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0028 - mae: 0.0442 - val_loss: 0.0023 - val_mae: 0.0384\nEpoch 121/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0030 - mae: 0.0442 - val_loss: 0.0023 - val_mae: 0.0383\nEpoch 122/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0032 - mae: 0.0449 - val_loss: 0.0023 - val_mae: 0.0390\nEpoch 123/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0033 - mae: 0.0477 - val_loss: 0.0025 - val_mae: 0.0407\nEpoch 124/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0029 - mae: 0.0447 - val_loss: 0.0024 - val_mae: 0.0397\nEpoch 125/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0025 - mae: 0.0425 - val_loss: 0.0023 - val_mae: 0.0409\nEpoch 126/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0028 - mae: 0.0453 - val_loss: 0.0027 - val_mae: 0.0444\nEpoch 127/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0036 - mae: 0.0527 - val_loss: 0.0025 - val_mae: 0.0415\nEpoch 128/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0029 - mae: 0.0452 - val_loss: 0.0023 - val_mae: 0.0387\nEpoch 129/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0445 - val_loss: 0.0026 - val_mae: 0.0424\nEpoch 130/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0027 - mae: 0.0463 - val_loss: 0.0032 - val_mae: 0.0533\nEpoch 131/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0032 - mae: 0.0515 - val_loss: 0.0023 - val_mae: 0.0387\nEpoch 132/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0029 - mae: 0.0460 - val_loss: 0.0023 - val_mae: 0.0383\nEpoch 133/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0027 - mae: 0.0448 - val_loss: 0.0023 - val_mae: 0.0383\nEpoch 134/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0030 - mae: 0.0439 - val_loss: 0.0023 - val_mae: 0.0406\nEpoch 135/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0028 - mae: 0.0427 - val_loss: 0.0023 - val_mae: 0.0382\nEpoch 136/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0032 - mae: 0.0462 - val_loss: 0.0023 - val_mae: 0.0412\nEpoch 137/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0029 - mae: 0.0439 - val_loss: 0.0024 - val_mae: 0.0396\nEpoch 138/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0027 - mae: 0.0435 - val_loss: 0.0023 - val_mae: 0.0390\nEpoch 139/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0438 - val_loss: 0.0025 - val_mae: 0.0408\nEpoch 140/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0029 - mae: 0.0473 - val_loss: 0.0023 - val_mae: 0.0382\nEpoch 141/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0027 - mae: 0.0432 - val_loss: 0.0025 - val_mae: 0.0453\nEpoch 142/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0030 - mae: 0.0456 - val_loss: 0.0023 - val_mae: 0.0387\nEpoch 143/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0026 - mae: 0.0422 - val_loss: 0.0024 - val_mae: 0.0393\nEpoch 144/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0035 - mae: 0.0495 - val_loss: 0.0025 - val_mae: 0.0411\nEpoch 145/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0033 - mae: 0.0578 - val_loss: 0.0023 - val_mae: 0.0387\nEpoch 146/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0033 - mae: 0.0508 - val_loss: 0.0023 - val_mae: 0.0384\nEpoch 147/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0452 - val_loss: 0.0023 - val_mae: 0.0384\nEpoch 148/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0029 - mae: 0.0453 - val_loss: 0.0025 - val_mae: 0.0409\nEpoch 149/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0027 - mae: 0.0448 - val_loss: 0.0023 - val_mae: 0.0384\nEpoch 150/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0028 - mae: 0.0435 - val_loss: 0.0025 - val_mae: 0.0477\nEpoch 151/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0031 - mae: 0.0518 - val_loss: 0.0023 - val_mae: 0.0384\nEpoch 152/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0455 - val_loss: 0.0023 - val_mae: 0.0384\nEpoch 153/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0026 - mae: 0.0459 - val_loss: 0.0024 - val_mae: 0.0438\nEpoch 154/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0025 - mae: 0.0454 - val_loss: 0.0023 - val_mae: 0.0388\nEpoch 155/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0033 - mae: 0.0487 - val_loss: 0.0023 - val_mae: 0.0386\nEpoch 156/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0028 - mae: 0.0438 - val_loss: 0.0024 - val_mae: 0.0423\nEpoch 157/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0026 - mae: 0.0430 - val_loss: 0.0023 - val_mae: 0.0390\nEpoch 158/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0445 - val_loss: 0.0031 - val_mae: 0.0528\nEpoch 159/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0035 - mae: 0.0515 - val_loss: 0.0023 - val_mae: 0.0385\nEpoch 160/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0031 - mae: 0.0446 - val_loss: 0.0023 - val_mae: 0.0395\nEpoch 161/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0032 - mae: 0.0503 - val_loss: 0.0023 - val_mae: 0.0385\nEpoch 162/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0028 - mae: 0.0452 - val_loss: 0.0026 - val_mae: 0.0494\nEpoch 163/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 8ms/step - loss: 0.0028 - mae: 0.0458 - val_loss: 0.0023 - val_mae: 0.0383\nEpoch 164/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0027 - mae: 0.0431 - val_loss: 0.0023 - val_mae: 0.0380\nEpoch 165/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0031 - mae: 0.0501 - val_loss: 0.0026 - val_mae: 0.0428\nEpoch 166/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0470 - val_loss: 0.0025 - val_mae: 0.0406\nEpoch 167/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0028 - mae: 0.0459 - val_loss: 0.0025 - val_mae: 0.0411\nEpoch 168/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0031 - mae: 0.0463 - val_loss: 0.0024 - val_mae: 0.0423\nEpoch 169/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0031 - mae: 0.0478 - val_loss: 0.0023 - val_mae: 0.0400\nEpoch 170/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0026 - mae: 0.0441 - val_loss: 0.0022 - val_mae: 0.0383\nEpoch 171/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0023 - mae: 0.0409 - val_loss: 0.0023 - val_mae: 0.0385\nEpoch 172/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0026 - mae: 0.0425 - val_loss: 0.0023 - val_mae: 0.0382\nEpoch 173/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0031 - mae: 0.0468 - val_loss: 0.0023 - val_mae: 0.0382\nEpoch 174/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 0.0031 - mae: 0.0466 - val_loss: 0.0051 - val_mae: 0.0804\nEpoch 175/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0049 - mae: 0.0727 - val_loss: 0.0024 - val_mae: 0.0430\nEpoch 176/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0447 - val_loss: 0.0024 - val_mae: 0.0394\nEpoch 177/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0034 - mae: 0.0476 - val_loss: 0.0024 - val_mae: 0.0393\nEpoch 178/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0464 - val_loss: 0.0023 - val_mae: 0.0409\nEpoch 179/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0454 - val_loss: 0.0026 - val_mae: 0.0429\nEpoch 180/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0024 - mae: 0.0439 - val_loss: 0.0034 - val_mae: 0.0569\nEpoch 181/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0034 - mae: 0.0565 - val_loss: 0.0023 - val_mae: 0.0380\nEpoch 182/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0032 - mae: 0.0464 - val_loss: 0.0023 - val_mae: 0.0390\nEpoch 183/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0463 - val_loss: 0.0023 - val_mae: 0.0414\nEpoch 184/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0028 - mae: 0.0449 - val_loss: 0.0024 - val_mae: 0.0399\nEpoch 185/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0031 - mae: 0.0473 - val_loss: 0.0023 - val_mae: 0.0378\nEpoch 186/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0455 - val_loss: 0.0023 - val_mae: 0.0390\nEpoch 187/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0028 - mae: 0.0453 - val_loss: 0.0023 - val_mae: 0.0387\nEpoch 188/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0032 - mae: 0.0480 - val_loss: 0.0024 - val_mae: 0.0445\nEpoch 189/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0455 - val_loss: 0.0024 - val_mae: 0.0402\nEpoch 190/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0030 - mae: 0.0455 - val_loss: 0.0023 - val_mae: 0.0383\nEpoch 191/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0029 - mae: 0.0432 - val_loss: 0.0024 - val_mae: 0.0426\nEpoch 192/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0028 - mae: 0.0431 - val_loss: 0.0023 - val_mae: 0.0383\nEpoch 193/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0035 - mae: 0.0548 - val_loss: 0.0024 - val_mae: 0.0402\nEpoch 194/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0424 - val_loss: 0.0023 - val_mae: 0.0391\nEpoch 195/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0032 - mae: 0.0456 - val_loss: 0.0023 - val_mae: 0.0405\nEpoch 196/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0028 - mae: 0.0418 - val_loss: 0.0023 - val_mae: 0.0382\nEpoch 197/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0035 - mae: 0.0477 - val_loss: 0.0023 - val_mae: 0.0380\nEpoch 198/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0028 - mae: 0.0451 - val_loss: 0.0024 - val_mae: 0.0397\nEpoch 199/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0025 - mae: 0.0425 - val_loss: 0.0027 - val_mae: 0.0441\nEpoch 200/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0035 - mae: 0.0490 - val_loss: 0.0023 - val_mae: 0.0390\nEpoch 201/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0027 - mae: 0.0426 - val_loss: 0.0023 - val_mae: 0.0381\nEpoch 202/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0452 - val_loss: 0.0022 - val_mae: 0.0382\nEpoch 203/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0025 - mae: 0.0417 - val_loss: 0.0022 - val_mae: 0.0382\nEpoch 204/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0026 - mae: 0.0443 - val_loss: 0.0027 - val_mae: 0.0451\nEpoch 205/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0468 - val_loss: 0.0023 - val_mae: 0.0385\nEpoch 206/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0031 - mae: 0.0466 - val_loss: 0.0026 - val_mae: 0.0431\nEpoch 207/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0028 - mae: 0.0455 - val_loss: 0.0023 - val_mae: 0.0385\nEpoch 208/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0029 - mae: 0.0461 - val_loss: 0.0023 - val_mae: 0.0379\nEpoch 209/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0023 - mae: 0.0397 - val_loss: 0.0023 - val_mae: 0.0391\nEpoch 210/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 8ms/step - loss: 0.0027 - mae: 0.0456 - val_loss: 0.0024 - val_mae: 0.0434\nEpoch 211/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step - loss: 0.0033 - mae: 0.0515 - val_loss: 0.0023 - val_mae: 0.0395\nEpoch 212/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0027 - mae: 0.0446 - val_loss: 0.0023 - val_mae: 0.0390\nEpoch 213/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0456 - val_loss: 0.0027 - val_mae: 0.0445\nEpoch 214/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0033 - mae: 0.0463 - val_loss: 0.0023 - val_mae: 0.0382\nEpoch 215/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0029 - mae: 0.0435 - val_loss: 0.0024 - val_mae: 0.0448\nEpoch 216/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0031 - mae: 0.0453 - val_loss: 0.0025 - val_mae: 0.0416\nEpoch 217/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0033 - mae: 0.0534 - val_loss: 0.0031 - val_mae: 0.0520\nEpoch 218/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0035 - mae: 0.0546 - val_loss: 0.0023 - val_mae: 0.0415\nEpoch 219/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0029 - mae: 0.0462 - val_loss: 0.0023 - val_mae: 0.0382\nEpoch 220/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0031 - mae: 0.0469 - val_loss: 0.0023 - val_mae: 0.0387\nEpoch 221/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0025 - mae: 0.0421 - val_loss: 0.0023 - val_mae: 0.0395\nEpoch 222/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0028 - mae: 0.0436 - val_loss: 0.0029 - val_mae: 0.0489\nEpoch 223/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0028 - mae: 0.0468 - val_loss: 0.0023 - val_mae: 0.0421\nEpoch 224/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0027 - mae: 0.0434 - val_loss: 0.0023 - val_mae: 0.0385\nEpoch 225/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0030 - mae: 0.0471 - val_loss: 0.0030 - val_mae: 0.0501\nEpoch 226/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0029 - mae: 0.0477 - val_loss: 0.0027 - val_mae: 0.0511\nEpoch 227/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0031 - mae: 0.0492 - val_loss: 0.0023 - val_mae: 0.0387\nEpoch 228/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0031 - mae: 0.0477 - val_loss: 0.0023 - val_mae: 0.0380\nEpoch 229/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0033 - mae: 0.0477 - val_loss: 0.0027 - val_mae: 0.0436\nEpoch 230/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0034 - mae: 0.0518 - val_loss: 0.0023 - val_mae: 0.0386\nEpoch 231/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0030 - mae: 0.0456 - val_loss: 0.0023 - val_mae: 0.0381\nEpoch 232/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 7ms/step - loss: 0.0035 - mae: 0.0501 - val_loss: 0.0036 - val_mae: 0.0605\nEpoch 233/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - loss: 0.0039 - mae: 0.0600 - val_loss: 0.0023 - val_mae: 0.0389\nEpoch 234/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0028 - mae: 0.0482 - val_loss: 0.0022 - val_mae: 0.0383\nEpoch 235/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - loss: 0.0026 - mae: 0.0442 - val_loss: 0.0026 - val_mae: 0.0428\n","output_type":"stream"}]},{"cell_type":"code","source":"# Not a very big change, worse than the SARIMA in terms of performance.\n# Next we'll build a better RNN.","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:26:29.548625Z","iopub.execute_input":"2024-07-11T17:26:29.549088Z","iopub.status.idle":"2024-07-11T17:26:29.555981Z","shell.execute_reply.started":"2024-07-11T17:26:29.549053Z","shell.execute_reply":"2024-07-11T17:26:29.554238Z"},"trusted":true},"execution_count":24,"outputs":[]},{"cell_type":"markdown","source":"# Forecasting using a simple RNN.","metadata":{}},{"cell_type":"code","source":"# We'll start with the most basic RNN containing a single recurrent layer with just one recurrent neuron.\n\nfrom tensorflow.keras.models import Sequential\nfrom tensorflow.keras.layers import SimpleRNN, Input, Dense\n\ntf.random.set_seed(42)\n\nunivar_model = tf.keras.Sequential([\n    Input(shape=(None, 1)),\n    SimpleRNN(32, return_sequences=False),\n    Dense(1) # No activation function for the regression task.\n])","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:41:03.159813Z","iopub.execute_input":"2024-07-11T17:41:03.160671Z","iopub.status.idle":"2024-07-11T17:41:03.242284Z","shell.execute_reply.started":"2024-07-11T17:41:03.160633Z","shell.execute_reply":"2024-07-11T17:41:03.240020Z"},"trusted":true},"execution_count":29,"outputs":[]},{"cell_type":"code","source":"# fitting and evaluationnutility function\n\ndef fit_and_evaluate(model, train_set, valid_set, learning_rate, epochs=500):\n    early_stopping_cb = tf.keras.callbacks.EarlyStopping(\n    monitor=\"val_mae\", patience=50, restore_best_weights=True)\n    opt = tf.keras.optimizers.SGD(learning_rate=learning_rate, momentum=0.9)\n    model.compile(loss=tf.keras.losses.Huber(), optimizer=opt, metrics=[\"mae\"])\n    history = model.fit(train_set, validation_data=valid_set, epochs=epochs, callbacks=[early_stopping_cb])\n    valid_loss, valid_mae = model.evaluate(valid_set)\n    return valid_mae * 1e6","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:46:36.465254Z","iopub.execute_input":"2024-07-11T17:46:36.465730Z","iopub.status.idle":"2024-07-11T17:46:36.475606Z","shell.execute_reply.started":"2024-07-11T17:46:36.465685Z","shell.execute_reply":"2024-07-11T17:46:36.474201Z"},"trusted":true},"execution_count":32,"outputs":[]},{"cell_type":"code","source":"fit_and_evaluate(univar_model, train_ds, valid_ds, learning_rate=0.05)","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:46:38.652500Z","iopub.execute_input":"2024-07-11T17:46:38.653241Z","iopub.status.idle":"2024-07-11T17:48:26.075545Z","shell.execute_reply.started":"2024-07-11T17:46:38.653196Z","shell.execute_reply":"2024-07-11T17:48:26.073814Z"},"trusted":true},"execution_count":33,"outputs":[{"name":"stdout","text":"Epoch 1/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 32ms/step - loss: 0.0241 - mae: 0.1762 - val_loss: 0.0035 - val_mae: 0.0626\nEpoch 2/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0049 - mae: 0.0719 - val_loss: 0.0027 - val_mae: 0.0526\nEpoch 3/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0044 - mae: 0.0627 - val_loss: 0.0038 - val_mae: 0.0697\nEpoch 4/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0043 - mae: 0.0604 - val_loss: 0.0033 - val_mae: 0.0538\nEpoch 5/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0038 - mae: 0.0544 - val_loss: 0.0022 - val_mae: 0.0398\nEpoch 6/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0031 - mae: 0.0521 - val_loss: 0.0028 - val_mae: 0.0495\nEpoch 7/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0031 - mae: 0.0511 - val_loss: 0.0021 - val_mae: 0.0379\nEpoch 8/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0033 - mae: 0.0496 - val_loss: 0.0026 - val_mae: 0.0473\nEpoch 9/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0032 - mae: 0.0508 - val_loss: 0.0021 - val_mae: 0.0368\nEpoch 10/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0029 - mae: 0.0499 - val_loss: 0.0024 - val_mae: 0.0463\nEpoch 11/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0034 - mae: 0.0497 - val_loss: 0.0021 - val_mae: 0.0359\nEpoch 12/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0032 - mae: 0.0487 - val_loss: 0.0031 - val_mae: 0.0543\nEpoch 13/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0030 - mae: 0.0495 - val_loss: 0.0023 - val_mae: 0.0375\nEpoch 14/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0030 - mae: 0.0473 - val_loss: 0.0023 - val_mae: 0.0358\nEpoch 15/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0036 - mae: 0.0514 - val_loss: 0.0020 - val_mae: 0.0359\nEpoch 16/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0034 - mae: 0.0466 - val_loss: 0.0026 - val_mae: 0.0502\nEpoch 17/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0031 - mae: 0.0496 - val_loss: 0.0030 - val_mae: 0.0543\nEpoch 18/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0038 - mae: 0.0547 - val_loss: 0.0019 - val_mae: 0.0322\nEpoch 19/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0027 - mae: 0.0432 - val_loss: 0.0021 - val_mae: 0.0356\nEpoch 20/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0031 - mae: 0.0482 - val_loss: 0.0021 - val_mae: 0.0347\nEpoch 21/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0029 - mae: 0.0441 - val_loss: 0.0019 - val_mae: 0.0331\nEpoch 22/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0028 - mae: 0.0470 - val_loss: 0.0022 - val_mae: 0.0404\nEpoch 23/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0035 - mae: 0.0541 - val_loss: 0.0020 - val_mae: 0.0358\nEpoch 24/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0026 - mae: 0.0425 - val_loss: 0.0022 - val_mae: 0.0386\nEpoch 25/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0031 - mae: 0.0482 - val_loss: 0.0021 - val_mae: 0.0377\nEpoch 26/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0023 - mae: 0.0413 - val_loss: 0.0019 - val_mae: 0.0342\nEpoch 27/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0032 - mae: 0.0483 - val_loss: 0.0032 - val_mae: 0.0585\nEpoch 28/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0036 - mae: 0.0568 - val_loss: 0.0026 - val_mae: 0.0485\nEpoch 29/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0036 - mae: 0.0509 - val_loss: 0.0020 - val_mae: 0.0352\nEpoch 30/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0030 - mae: 0.0478 - val_loss: 0.0019 - val_mae: 0.0339\nEpoch 31/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0024 - mae: 0.0425 - val_loss: 0.0020 - val_mae: 0.0368\nEpoch 32/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0027 - mae: 0.0426 - val_loss: 0.0018 - val_mae: 0.0326\nEpoch 33/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0028 - mae: 0.0441 - val_loss: 0.0020 - val_mae: 0.0377\nEpoch 34/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0028 - mae: 0.0448 - val_loss: 0.0021 - val_mae: 0.0380\nEpoch 35/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0026 - mae: 0.0428 - val_loss: 0.0025 - val_mae: 0.0493\nEpoch 36/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0026 - mae: 0.0458 - val_loss: 0.0024 - val_mae: 0.0454\nEpoch 37/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0028 - mae: 0.0453 - val_loss: 0.0019 - val_mae: 0.0338\nEpoch 38/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0026 - mae: 0.0413 - val_loss: 0.0023 - val_mae: 0.0422\nEpoch 39/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0029 - mae: 0.0427 - val_loss: 0.0019 - val_mae: 0.0345\nEpoch 40/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0029 - mae: 0.0444 - val_loss: 0.0019 - val_mae: 0.0347\nEpoch 41/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0027 - mae: 0.0462 - val_loss: 0.0021 - val_mae: 0.0373\nEpoch 42/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0458 - val_loss: 0.0021 - val_mae: 0.0419\nEpoch 43/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0029 - mae: 0.0447 - val_loss: 0.0022 - val_mae: 0.0408\nEpoch 44/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0034 - mae: 0.0507 - val_loss: 0.0020 - val_mae: 0.0368\nEpoch 45/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0031 - mae: 0.0446 - val_loss: 0.0021 - val_mae: 0.0388\nEpoch 46/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0461 - val_loss: 0.0024 - val_mae: 0.0457\nEpoch 47/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0028 - mae: 0.0455 - val_loss: 0.0021 - val_mae: 0.0375\nEpoch 48/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0034 - mae: 0.0502 - val_loss: 0.0021 - val_mae: 0.0397\nEpoch 49/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0025 - mae: 0.0423 - val_loss: 0.0022 - val_mae: 0.0428\nEpoch 50/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0027 - mae: 0.0435 - val_loss: 0.0030 - val_mae: 0.0567\nEpoch 51/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0027 - mae: 0.0450 - val_loss: 0.0019 - val_mae: 0.0363\nEpoch 52/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0026 - mae: 0.0429 - val_loss: 0.0019 - val_mae: 0.0365\nEpoch 53/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0028 - mae: 0.0407 - val_loss: 0.0019 - val_mae: 0.0366\nEpoch 54/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0025 - mae: 0.0421 - val_loss: 0.0019 - val_mae: 0.0316\nEpoch 55/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0029 - mae: 0.0434 - val_loss: 0.0028 - val_mae: 0.0521\nEpoch 56/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0033 - mae: 0.0476 - val_loss: 0.0019 - val_mae: 0.0318\nEpoch 57/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0027 - mae: 0.0430 - val_loss: 0.0020 - val_mae: 0.0372\nEpoch 58/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0026 - mae: 0.0408 - val_loss: 0.0019 - val_mae: 0.0328\nEpoch 59/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0025 - mae: 0.0411 - val_loss: 0.0020 - val_mae: 0.0382\nEpoch 60/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0028 - mae: 0.0451 - val_loss: 0.0020 - val_mae: 0.0378\nEpoch 61/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0030 - mae: 0.0447 - val_loss: 0.0019 - val_mae: 0.0365\nEpoch 62/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0030 - mae: 0.0463 - val_loss: 0.0020 - val_mae: 0.0343\nEpoch 63/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0426 - val_loss: 0.0018 - val_mae: 0.0333\nEpoch 64/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0029 - mae: 0.0455 - val_loss: 0.0022 - val_mae: 0.0407\nEpoch 65/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0022 - mae: 0.0421 - val_loss: 0.0021 - val_mae: 0.0392\nEpoch 66/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0024 - mae: 0.0428 - val_loss: 0.0021 - val_mae: 0.0389\nEpoch 67/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0024 - mae: 0.0407 - val_loss: 0.0020 - val_mae: 0.0365\nEpoch 68/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0025 - mae: 0.0412 - val_loss: 0.0018 - val_mae: 0.0313\nEpoch 69/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0028 - mae: 0.0405 - val_loss: 0.0019 - val_mae: 0.0361\nEpoch 70/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0027 - mae: 0.0442 - val_loss: 0.0018 - val_mae: 0.0328\nEpoch 71/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0029 - mae: 0.0431 - val_loss: 0.0031 - val_mae: 0.0593\nEpoch 72/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0035 - mae: 0.0520 - val_loss: 0.0021 - val_mae: 0.0421\nEpoch 73/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0026 - mae: 0.0416 - val_loss: 0.0020 - val_mae: 0.0369\nEpoch 74/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0023 - mae: 0.0406 - val_loss: 0.0021 - val_mae: 0.0399\nEpoch 75/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0026 - mae: 0.0412 - val_loss: 0.0025 - val_mae: 0.0475\nEpoch 76/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0024 - mae: 0.0422 - val_loss: 0.0019 - val_mae: 0.0351\nEpoch 77/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0029 - mae: 0.0450 - val_loss: 0.0019 - val_mae: 0.0327\nEpoch 78/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0029 - mae: 0.0417 - val_loss: 0.0018 - val_mae: 0.0315\nEpoch 79/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0027 - mae: 0.0427 - val_loss: 0.0020 - val_mae: 0.0393\nEpoch 80/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0026 - mae: 0.0415 - val_loss: 0.0023 - val_mae: 0.0452\nEpoch 81/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0024 - mae: 0.0422 - val_loss: 0.0020 - val_mae: 0.0381\nEpoch 82/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0025 - mae: 0.0418 - val_loss: 0.0018 - val_mae: 0.0302\nEpoch 83/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0028 - mae: 0.0421 - val_loss: 0.0018 - val_mae: 0.0318\nEpoch 84/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0028 - mae: 0.0433 - val_loss: 0.0018 - val_mae: 0.0333\nEpoch 85/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0022 - mae: 0.0388 - val_loss: 0.0019 - val_mae: 0.0354\nEpoch 86/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0028 - mae: 0.0451 - val_loss: 0.0018 - val_mae: 0.0309\nEpoch 87/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0028 - mae: 0.0420 - val_loss: 0.0022 - val_mae: 0.0420\nEpoch 88/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0028 - mae: 0.0444 - val_loss: 0.0017 - val_mae: 0.0292\nEpoch 89/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0025 - mae: 0.0419 - val_loss: 0.0018 - val_mae: 0.0320\nEpoch 90/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0025 - mae: 0.0407 - val_loss: 0.0019 - val_mae: 0.0352\nEpoch 91/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0028 - mae: 0.0449 - val_loss: 0.0018 - val_mae: 0.0322\nEpoch 92/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0026 - mae: 0.0402 - val_loss: 0.0020 - val_mae: 0.0395\nEpoch 93/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0025 - mae: 0.0412 - val_loss: 0.0020 - val_mae: 0.0382\nEpoch 94/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0032 - mae: 0.0461 - val_loss: 0.0018 - val_mae: 0.0309\nEpoch 95/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0026 - mae: 0.0417 - val_loss: 0.0017 - val_mae: 0.0329\nEpoch 96/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0027 - mae: 0.0416 - val_loss: 0.0019 - val_mae: 0.0343\nEpoch 97/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0028 - mae: 0.0410 - val_loss: 0.0018 - val_mae: 0.0327\nEpoch 98/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0021 - mae: 0.0379 - val_loss: 0.0018 - val_mae: 0.0288\nEpoch 99/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0026 - mae: 0.0394 - val_loss: 0.0019 - val_mae: 0.0344\nEpoch 100/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0027 - mae: 0.0412 - val_loss: 0.0020 - val_mae: 0.0361\nEpoch 101/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0029 - mae: 0.0416 - val_loss: 0.0018 - val_mae: 0.0338\nEpoch 102/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0028 - mae: 0.0422 - val_loss: 0.0017 - val_mae: 0.0307\nEpoch 103/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0026 - mae: 0.0411 - val_loss: 0.0018 - val_mae: 0.0339\nEpoch 104/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0029 - mae: 0.0440 - val_loss: 0.0018 - val_mae: 0.0338\nEpoch 105/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0026 - mae: 0.0394 - val_loss: 0.0022 - val_mae: 0.0423\nEpoch 106/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0025 - mae: 0.0430 - val_loss: 0.0021 - val_mae: 0.0412\nEpoch 107/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0025 - mae: 0.0408 - val_loss: 0.0018 - val_mae: 0.0317\nEpoch 108/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0023 - mae: 0.0400 - val_loss: 0.0019 - val_mae: 0.0352\nEpoch 109/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0023 - mae: 0.0387 - val_loss: 0.0019 - val_mae: 0.0328\nEpoch 110/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0031 - mae: 0.0476 - val_loss: 0.0019 - val_mae: 0.0362\nEpoch 111/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0030 - mae: 0.0478 - val_loss: 0.0018 - val_mae: 0.0315\nEpoch 112/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0031 - mae: 0.0444 - val_loss: 0.0018 - val_mae: 0.0301\nEpoch 113/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0025 - mae: 0.0398 - val_loss: 0.0018 - val_mae: 0.0340\nEpoch 114/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0024 - mae: 0.0406 - val_loss: 0.0018 - val_mae: 0.0350\nEpoch 115/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0028 - mae: 0.0433 - val_loss: 0.0018 - val_mae: 0.0345\nEpoch 116/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0025 - mae: 0.0419 - val_loss: 0.0019 - val_mae: 0.0333\nEpoch 117/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0031 - mae: 0.0446 - val_loss: 0.0022 - val_mae: 0.0445\nEpoch 118/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0031 - mae: 0.0432 - val_loss: 0.0018 - val_mae: 0.0315\nEpoch 119/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0026 - mae: 0.0408 - val_loss: 0.0022 - val_mae: 0.0444\nEpoch 120/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0025 - mae: 0.0422 - val_loss: 0.0020 - val_mae: 0.0358\nEpoch 121/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0030 - mae: 0.0426 - val_loss: 0.0018 - val_mae: 0.0328\nEpoch 122/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0025 - mae: 0.0409 - val_loss: 0.0018 - val_mae: 0.0352\nEpoch 123/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0021 - mae: 0.0374 - val_loss: 0.0017 - val_mae: 0.0308\nEpoch 124/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0029 - mae: 0.0445 - val_loss: 0.0025 - val_mae: 0.0457\nEpoch 125/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0033 - mae: 0.0488 - val_loss: 0.0017 - val_mae: 0.0325\nEpoch 126/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0026 - mae: 0.0427 - val_loss: 0.0018 - val_mae: 0.0335\nEpoch 127/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0024 - mae: 0.0403 - val_loss: 0.0022 - val_mae: 0.0445\nEpoch 128/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0026 - mae: 0.0428 - val_loss: 0.0020 - val_mae: 0.0365\nEpoch 129/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0028 - mae: 0.0435 - val_loss: 0.0018 - val_mae: 0.0352\nEpoch 130/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0033 - mae: 0.0470 - val_loss: 0.0018 - val_mae: 0.0341\nEpoch 131/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0023 - mae: 0.0381 - val_loss: 0.0018 - val_mae: 0.0332\nEpoch 132/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0027 - mae: 0.0416 - val_loss: 0.0018 - val_mae: 0.0325\nEpoch 133/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0030 - mae: 0.0462 - val_loss: 0.0019 - val_mae: 0.0354\nEpoch 134/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0025 - mae: 0.0408 - val_loss: 0.0020 - val_mae: 0.0361\nEpoch 135/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0022 - mae: 0.0359 - val_loss: 0.0021 - val_mae: 0.0433\nEpoch 136/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0031 - mae: 0.0471 - val_loss: 0.0018 - val_mae: 0.0305\nEpoch 137/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0025 - mae: 0.0403 - val_loss: 0.0020 - val_mae: 0.0392\nEpoch 138/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0029 - mae: 0.0472 - val_loss: 0.0018 - val_mae: 0.0346\nEpoch 139/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0024 - mae: 0.0389 - val_loss: 0.0017 - val_mae: 0.0321\nEpoch 140/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0026 - mae: 0.0442 - val_loss: 0.0019 - val_mae: 0.0347\nEpoch 141/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0026 - mae: 0.0422 - val_loss: 0.0018 - val_mae: 0.0330\nEpoch 142/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0023 - mae: 0.0383 - val_loss: 0.0018 - val_mae: 0.0347\nEpoch 143/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0025 - mae: 0.0411 - val_loss: 0.0023 - val_mae: 0.0447\nEpoch 144/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0027 - mae: 0.0455 - val_loss: 0.0019 - val_mae: 0.0358\nEpoch 145/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0025 - mae: 0.0401 - val_loss: 0.0019 - val_mae: 0.0360\nEpoch 146/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0026 - mae: 0.0449 - val_loss: 0.0019 - val_mae: 0.0341\nEpoch 147/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0026 - mae: 0.0413 - val_loss: 0.0018 - val_mae: 0.0328\nEpoch 148/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0026 - mae: 0.0400 - val_loss: 0.0018 - val_mae: 0.0344\n\u001b[1m3/3\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 9ms/step - loss: 0.0013 - mae: 0.0264     \n","output_type":"stream"},{"execution_count":33,"output_type":"execute_result","data":{"text/plain":"28787.827119231224"},"metadata":{}}]},{"cell_type":"code","source":"# This is a much better result, even without removing the trend and seasonality.\n# However for best performance we might want to try making time series more stationary e.g using differencing.","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:50:03.665068Z","iopub.execute_input":"2024-07-11T17:50:03.665575Z","iopub.status.idle":"2024-07-11T17:50:03.671808Z","shell.execute_reply.started":"2024-07-11T17:50:03.665541Z","shell.execute_reply":"2024-07-11T17:50:03.670269Z"},"trusted":true},"execution_count":34,"outputs":[]},{"cell_type":"markdown","source":"# Forecasting using a Deep Recurrent Neural Network.","metadata":{}},{"cell_type":"code","source":"from tensorflow.keras.optimizers import Adam\n\ndeep_model = Sequential([\n    Input(shape=(None, 1)),\n    SimpleRNN(32, return_sequences=True),\n    SimpleRNN(32, return_sequences=True),\n    SimpleRNN(32),\n    Dense(1)\n])","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:56:18.346118Z","iopub.execute_input":"2024-07-11T17:56:18.346622Z","iopub.status.idle":"2024-07-11T17:56:18.464870Z","shell.execute_reply.started":"2024-07-11T17:56:18.346585Z","shell.execute_reply":"2024-07-11T17:56:18.461844Z"},"trusted":true},"execution_count":36,"outputs":[]},{"cell_type":"code","source":"fit_and_evaluate(deep_model, train_ds, valid_ds, learning_rate=0.01)","metadata":{"execution":{"iopub.status.busy":"2024-07-11T17:57:35.772640Z","iopub.execute_input":"2024-07-11T17:57:35.773227Z","iopub.status.idle":"2024-07-11T18:00:17.166720Z","shell.execute_reply.started":"2024-07-11T17:57:35.773187Z","shell.execute_reply":"2024-07-11T18:00:17.164851Z"},"trusted":true},"execution_count":37,"outputs":[{"name":"stdout","text":"Epoch 1/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 78ms/step - loss: 0.0736 - mae: 0.2632 - val_loss: 0.0060 - val_mae: 0.0849\nEpoch 2/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 56ms/step - loss: 0.0072 - mae: 0.0857 - val_loss: 0.0026 - val_mae: 0.0523\nEpoch 3/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 54ms/step - loss: 0.0053 - mae: 0.0709 - val_loss: 0.0030 - val_mae: 0.0563\nEpoch 4/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 51ms/step - loss: 0.0043 - mae: 0.0625 - val_loss: 0.0024 - val_mae: 0.0467\nEpoch 5/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 51ms/step - loss: 0.0042 - mae: 0.0615 - val_loss: 0.0037 - val_mae: 0.0635\nEpoch 6/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - loss: 0.0042 - mae: 0.0615 - val_loss: 0.0020 - val_mae: 0.0392\nEpoch 7/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - loss: 0.0043 - mae: 0.0586 - val_loss: 0.0039 - val_mae: 0.0679\nEpoch 8/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 55ms/step - loss: 0.0041 - mae: 0.0574 - val_loss: 0.0030 - val_mae: 0.0538\nEpoch 9/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - loss: 0.0036 - mae: 0.0519 - val_loss: 0.0026 - val_mae: 0.0475\nEpoch 10/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 52ms/step - loss: 0.0042 - mae: 0.0614 - val_loss: 0.0020 - val_mae: 0.0368\nEpoch 11/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - loss: 0.0037 - mae: 0.0515 - val_loss: 0.0024 - val_mae: 0.0426\nEpoch 12/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - loss: 0.0029 - mae: 0.0476 - val_loss: 0.0020 - val_mae: 0.0371\nEpoch 13/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 46ms/step - loss: 0.0041 - mae: 0.0563 - val_loss: 0.0020 - val_mae: 0.0362\nEpoch 14/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 50ms/step - loss: 0.0030 - mae: 0.0461 - val_loss: 0.0019 - val_mae: 0.0319\nEpoch 15/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 48ms/step - loss: 0.0032 - mae: 0.0477 - val_loss: 0.0020 - val_mae: 0.0360\nEpoch 16/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 52ms/step - loss: 0.0039 - mae: 0.0541 - val_loss: 0.0020 - val_mae: 0.0331\nEpoch 17/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 54ms/step - loss: 0.0031 - mae: 0.0482 - val_loss: 0.0021 - val_mae: 0.0345\nEpoch 18/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 53ms/step - loss: 0.0034 - mae: 0.0501 - val_loss: 0.0021 - val_mae: 0.0363\nEpoch 19/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 51ms/step - loss: 0.0033 - mae: 0.0469 - val_loss: 0.0024 - val_mae: 0.0404\nEpoch 20/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 51ms/step - loss: 0.0035 - mae: 0.0475 - val_loss: 0.0026 - val_mae: 0.0449\nEpoch 21/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - loss: 0.0034 - mae: 0.0487 - val_loss: 0.0020 - val_mae: 0.0329\nEpoch 22/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 48ms/step - loss: 0.0030 - mae: 0.0467 - val_loss: 0.0018 - val_mae: 0.0335\nEpoch 23/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 51ms/step - loss: 0.0034 - mae: 0.0495 - val_loss: 0.0019 - val_mae: 0.0327\nEpoch 24/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 50ms/step - loss: 0.0031 - mae: 0.0485 - val_loss: 0.0035 - val_mae: 0.0608\nEpoch 25/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - loss: 0.0035 - mae: 0.0511 - val_loss: 0.0020 - val_mae: 0.0341\nEpoch 26/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - loss: 0.0027 - mae: 0.0457 - val_loss: 0.0021 - val_mae: 0.0343\nEpoch 27/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 48ms/step - loss: 0.0033 - mae: 0.0483 - val_loss: 0.0021 - val_mae: 0.0365\nEpoch 28/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 55ms/step - loss: 0.0030 - mae: 0.0451 - val_loss: 0.0029 - val_mae: 0.0557\nEpoch 29/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 51ms/step - loss: 0.0028 - mae: 0.0484 - val_loss: 0.0020 - val_mae: 0.0353\nEpoch 30/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 51ms/step - loss: 0.0026 - mae: 0.0430 - val_loss: 0.0019 - val_mae: 0.0345\nEpoch 31/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 48ms/step - loss: 0.0029 - mae: 0.0461 - val_loss: 0.0019 - val_mae: 0.0361\nEpoch 32/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 48ms/step - loss: 0.0027 - mae: 0.0463 - val_loss: 0.0018 - val_mae: 0.0306\nEpoch 33/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 54ms/step - loss: 0.0027 - mae: 0.0440 - val_loss: 0.0018 - val_mae: 0.0294\nEpoch 34/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 51ms/step - loss: 0.0028 - mae: 0.0422 - val_loss: 0.0029 - val_mae: 0.0534\nEpoch 35/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - loss: 0.0033 - mae: 0.0500 - val_loss: 0.0018 - val_mae: 0.0294\nEpoch 36/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - loss: 0.0029 - mae: 0.0444 - val_loss: 0.0021 - val_mae: 0.0384\nEpoch 37/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - loss: 0.0030 - mae: 0.0485 - val_loss: 0.0018 - val_mae: 0.0299\nEpoch 38/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 50ms/step - loss: 0.0028 - mae: 0.0444 - val_loss: 0.0018 - val_mae: 0.0302\nEpoch 39/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - loss: 0.0030 - mae: 0.0445 - val_loss: 0.0021 - val_mae: 0.0387\nEpoch 40/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 48ms/step - loss: 0.0028 - mae: 0.0455 - val_loss: 0.0021 - val_mae: 0.0398\nEpoch 41/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - loss: 0.0029 - mae: 0.0454 - val_loss: 0.0019 - val_mae: 0.0358\nEpoch 42/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 50ms/step - loss: 0.0031 - mae: 0.0482 - val_loss: 0.0018 - val_mae: 0.0311\nEpoch 43/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 50ms/step - loss: 0.0026 - mae: 0.0428 - val_loss: 0.0018 - val_mae: 0.0308\nEpoch 44/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 52ms/step - loss: 0.0027 - mae: 0.0432 - val_loss: 0.0020 - val_mae: 0.0372\nEpoch 45/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 54ms/step - loss: 0.0025 - mae: 0.0421 - val_loss: 0.0018 - val_mae: 0.0308\nEpoch 46/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 53ms/step - loss: 0.0025 - mae: 0.0419 - val_loss: 0.0020 - val_mae: 0.0372\nEpoch 47/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - loss: 0.0031 - mae: 0.0505 - val_loss: 0.0020 - val_mae: 0.0402\nEpoch 48/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - loss: 0.0025 - mae: 0.0434 - val_loss: 0.0029 - val_mae: 0.0535\nEpoch 49/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 57ms/step - loss: 0.0028 - mae: 0.0456 - val_loss: 0.0021 - val_mae: 0.0423\nEpoch 50/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 55ms/step - loss: 0.0033 - mae: 0.0540 - val_loss: 0.0021 - val_mae: 0.0378\nEpoch 51/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 53ms/step - loss: 0.0031 - mae: 0.0496 - val_loss: 0.0018 - val_mae: 0.0331\nEpoch 52/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 55ms/step - loss: 0.0029 - mae: 0.0444 - val_loss: 0.0021 - val_mae: 0.0405\nEpoch 53/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 51ms/step - loss: 0.0026 - mae: 0.0432 - val_loss: 0.0018 - val_mae: 0.0310\nEpoch 54/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 52ms/step - loss: 0.0023 - mae: 0.0406 - val_loss: 0.0019 - val_mae: 0.0325\nEpoch 55/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 51ms/step - loss: 0.0028 - mae: 0.0465 - val_loss: 0.0018 - val_mae: 0.0306\nEpoch 56/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 52ms/step - loss: 0.0030 - mae: 0.0458 - val_loss: 0.0017 - val_mae: 0.0295\nEpoch 57/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 48ms/step - loss: 0.0028 - mae: 0.0434 - val_loss: 0.0018 - val_mae: 0.0326\nEpoch 58/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - loss: 0.0027 - mae: 0.0424 - val_loss: 0.0020 - val_mae: 0.0374\nEpoch 59/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 50ms/step - loss: 0.0028 - mae: 0.0445 - val_loss: 0.0018 - val_mae: 0.0316\nEpoch 60/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 53ms/step - loss: 0.0029 - mae: 0.0434 - val_loss: 0.0018 - val_mae: 0.0333\nEpoch 61/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - loss: 0.0025 - mae: 0.0395 - val_loss: 0.0018 - val_mae: 0.0308\nEpoch 62/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 52ms/step - loss: 0.0024 - mae: 0.0425 - val_loss: 0.0017 - val_mae: 0.0305\nEpoch 63/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 51ms/step - loss: 0.0024 - mae: 0.0400 - val_loss: 0.0018 - val_mae: 0.0314\nEpoch 64/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 47ms/step - loss: 0.0028 - mae: 0.0435 - val_loss: 0.0018 - val_mae: 0.0319\nEpoch 65/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 57ms/step - loss: 0.0030 - mae: 0.0446 - val_loss: 0.0020 - val_mae: 0.0355\nEpoch 66/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 60ms/step - loss: 0.0031 - mae: 0.0446 - val_loss: 0.0019 - val_mae: 0.0351\nEpoch 67/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 58ms/step - loss: 0.0029 - mae: 0.0449 - val_loss: 0.0021 - val_mae: 0.0370\nEpoch 68/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 58ms/step - loss: 0.0033 - mae: 0.0495 - val_loss: 0.0018 - val_mae: 0.0326\nEpoch 69/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 50ms/step - loss: 0.0024 - mae: 0.0403 - val_loss: 0.0025 - val_mae: 0.0514\nEpoch 70/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 57ms/step - loss: 0.0028 - mae: 0.0464 - val_loss: 0.0018 - val_mae: 0.0301\nEpoch 71/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 56ms/step - loss: 0.0027 - mae: 0.0425 - val_loss: 0.0019 - val_mae: 0.0367\nEpoch 72/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 50ms/step - loss: 0.0028 - mae: 0.0442 - val_loss: 0.0019 - val_mae: 0.0337\nEpoch 73/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 51ms/step - loss: 0.0030 - mae: 0.0484 - val_loss: 0.0019 - val_mae: 0.0391\nEpoch 74/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - loss: 0.0024 - mae: 0.0403 - val_loss: 0.0018 - val_mae: 0.0329\nEpoch 75/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 52ms/step - loss: 0.0029 - mae: 0.0462 - val_loss: 0.0018 - val_mae: 0.0326\nEpoch 76/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 49ms/step - loss: 0.0028 - mae: 0.0473 - val_loss: 0.0024 - val_mae: 0.0496\nEpoch 77/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 51ms/step - loss: 0.0030 - mae: 0.0473 - val_loss: 0.0020 - val_mae: 0.0392\nEpoch 78/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 50ms/step - loss: 0.0027 - mae: 0.0426 - val_loss: 0.0021 - val_mae: 0.0403\nEpoch 79/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 51ms/step - loss: 0.0030 - mae: 0.0470 - val_loss: 0.0018 - val_mae: 0.0333\nEpoch 80/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 51ms/step - loss: 0.0028 - mae: 0.0404 - val_loss: 0.0021 - val_mae: 0.0388\nEpoch 81/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 57ms/step - loss: 0.0029 - mae: 0.0451 - val_loss: 0.0018 - val_mae: 0.0325\nEpoch 82/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 58ms/step - loss: 0.0027 - mae: 0.0447 - val_loss: 0.0023 - val_mae: 0.0463\nEpoch 83/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 54ms/step - loss: 0.0029 - mae: 0.0444 - val_loss: 0.0017 - val_mae: 0.0298\nEpoch 84/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 53ms/step - loss: 0.0026 - mae: 0.0426 - val_loss: 0.0021 - val_mae: 0.0405\nEpoch 85/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 52ms/step - loss: 0.0024 - mae: 0.0439 - val_loss: 0.0018 - val_mae: 0.0315\n\u001b[1m3/3\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 17ms/step - loss: 0.0013 - mae: 0.0274    \n","output_type":"stream"},{"execution_count":37,"output_type":"execute_result","data":{"text/plain":"29386.26892864704"},"metadata":{}}]},{"cell_type":"code","source":"# This is worse than our previous model! The network is too large for our task.","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:01:32.077965Z","iopub.execute_input":"2024-07-11T18:01:32.078464Z","iopub.status.idle":"2024-07-11T18:01:32.085377Z","shell.execute_reply.started":"2024-07-11T18:01:32.078430Z","shell.execute_reply":"2024-07-11T18:01:32.083493Z"},"trusted":true},"execution_count":38,"outputs":[]},{"cell_type":"markdown","source":"# Forecasting Multivariate Time Series with Recurrent Neural Networks","metadata":{}},{"cell_type":"code","source":"# Lets try to forecast the rail time series using both the the bus and rail data as input - also the day type.\n\ndf_mulvar = df[[\"bus\", \"rail\"]] / 1e6 # use both the bus and rail series as inputs.\ndf_mulvar[\"next_day_type\"] = df[\"day_type\"].shift(-1) # we know tomorrow's type.\ndf_mulvar = pd.get_dummies(df_mulvar) # one-hot encode the day type.","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:07:19.855444Z","iopub.execute_input":"2024-07-11T18:07:19.855965Z","iopub.status.idle":"2024-07-11T18:07:19.876750Z","shell.execute_reply.started":"2024-07-11T18:07:19.855899Z","shell.execute_reply":"2024-07-11T18:07:19.874748Z"},"trusted":true},"execution_count":39,"outputs":[]},{"cell_type":"code","source":"bool_columns = ['next_day_type_A', 'next_day_type_U', 'next_day_type_W']\ndf_mulvar[bool_columns] = df_mulvar[bool_columns].astype(int)\n\n# Convert all data to float32\ndf_mulvar = df_mulvar.astype('float32')","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:16:59.897749Z","iopub.execute_input":"2024-07-11T18:16:59.899308Z","iopub.status.idle":"2024-07-11T18:16:59.910947Z","shell.execute_reply.started":"2024-07-11T18:16:59.899259Z","shell.execute_reply":"2024-07-11T18:16:59.909089Z"},"trusted":true},"execution_count":47,"outputs":[]},{"cell_type":"code","source":"df_mulvar.dtypes","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:17:21.084083Z","iopub.execute_input":"2024-07-11T18:17:21.084542Z","iopub.status.idle":"2024-07-11T18:17:21.095821Z","shell.execute_reply.started":"2024-07-11T18:17:21.084508Z","shell.execute_reply":"2024-07-11T18:17:21.093771Z"},"trusted":true},"execution_count":48,"outputs":[{"execution_count":48,"output_type":"execute_result","data":{"text/plain":"bus                float32\nrail               float32\nnext_day_type_A    float32\nnext_day_type_U    float32\nnext_day_type_W    float32\ndtype: object"},"metadata":{}}]},{"cell_type":"code","source":"# Next we split the data into three periods for training, validation and testing.\n\nmulvar_train = df_mulvar[\"2016-01\":\"2018-12\"]\nmulvar_valid = df_mulvar[\"2019-01\":\"2019-05\"]\nmulvar_test = df_mulvar[\"2019-06\":]","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:17:34.566761Z","iopub.execute_input":"2024-07-11T18:17:34.567330Z","iopub.status.idle":"2024-07-11T18:17:34.584622Z","shell.execute_reply.started":"2024-07-11T18:17:34.567288Z","shell.execute_reply":"2024-07-11T18:17:34.582325Z"},"trusted":true},"execution_count":49,"outputs":[]},{"cell_type":"code","source":"# Next we create the datasets.\n\ntf.random.set_seed(42)  # extra code – ensures reproducibility\n\ntrain_mulvar_ds = tf.keras.utils.timeseries_dataset_from_array(\n    mulvar_train.to_numpy(),  # use all 5 columns as input\n    targets=mulvar_train[\"rail\"][seq_length:],  # forecast only the rail series\n    sequence_length=seq_length,\n    batch_size=32,\n    shuffle=True,\n    seed=42\n)\nvalid_mulvar_ds = tf.keras.utils.timeseries_dataset_from_array(\n    mulvar_valid.to_numpy(),\n    targets=mulvar_valid[\"rail\"][seq_length:],\n    sequence_length=seq_length,\n    batch_size=32\n)","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:17:36.860999Z","iopub.execute_input":"2024-07-11T18:17:36.861517Z","iopub.status.idle":"2024-07-11T18:17:37.093468Z","shell.execute_reply.started":"2024-07-11T18:17:36.861480Z","shell.execute_reply":"2024-07-11T18:17:37.091702Z"},"trusted":true},"execution_count":50,"outputs":[]},{"cell_type":"code","source":"# Building the model.\n\nmulvar_model = Sequential([\n    Input(shape=(None, 5)),\n    SimpleRNN(32, return_sequences=False),\n    Dense(1)\n])","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:19:45.562532Z","iopub.execute_input":"2024-07-11T18:19:45.563180Z","iopub.status.idle":"2024-07-11T18:19:45.642249Z","shell.execute_reply.started":"2024-07-11T18:19:45.563132Z","shell.execute_reply":"2024-07-11T18:19:45.640364Z"},"trusted":true},"execution_count":52,"outputs":[]},{"cell_type":"code","source":"fit_and_evaluate(mulvar_model, train_mulvar_ds, valid_mulvar_ds, learning_rate=0.05)","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:20:40.033705Z","iopub.execute_input":"2024-07-11T18:20:40.034279Z","iopub.status.idle":"2024-07-11T18:22:04.171873Z","shell.execute_reply.started":"2024-07-11T18:20:40.034238Z","shell.execute_reply":"2024-07-11T18:22:04.170146Z"},"trusted":true},"execution_count":53,"outputs":[{"name":"stdout","text":"Epoch 1/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 29ms/step - loss: 0.0885 - mae: 0.3188 - val_loss: 0.0030 - val_mae: 0.0664\nEpoch 2/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0053 - mae: 0.0792 - val_loss: 0.0034 - val_mae: 0.0737\nEpoch 3/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0028 - mae: 0.0577 - val_loss: 7.5484e-04 - val_mae: 0.0285\nEpoch 4/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0019 - mae: 0.0462 - val_loss: 0.0017 - val_mae: 0.0488\nEpoch 5/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0017 - mae: 0.0417 - val_loss: 7.3849e-04 - val_mae: 0.0275\nEpoch 6/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0013 - mae: 0.0362 - val_loss: 0.0011 - val_mae: 0.0359\nEpoch 7/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0013 - mae: 0.0371 - val_loss: 8.4763e-04 - val_mae: 0.0306\nEpoch 8/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0014 - mae: 0.0369 - val_loss: 6.6945e-04 - val_mae: 0.0255\nEpoch 9/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0351 - val_loss: 7.6823e-04 - val_mae: 0.0291\nEpoch 10/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0012 - mae: 0.0352 - val_loss: 6.8581e-04 - val_mae: 0.0261\nEpoch 11/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0014 - mae: 0.0400 - val_loss: 9.2195e-04 - val_mae: 0.0329\nEpoch 12/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0328 - val_loss: 0.0010 - val_mae: 0.0356\nEpoch 13/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0345 - val_loss: 6.5820e-04 - val_mae: 0.0256\nEpoch 14/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0354 - val_loss: 0.0015 - val_mae: 0.0457\nEpoch 15/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0013 - mae: 0.0361 - val_loss: 6.0644e-04 - val_mae: 0.0240\nEpoch 16/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0345 - val_loss: 6.2972e-04 - val_mae: 0.0246\nEpoch 17/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0010 - mae: 0.0326 - val_loss: 9.6279e-04 - val_mae: 0.0340\nEpoch 18/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0335 - val_loss: 6.1733e-04 - val_mae: 0.0260\nEpoch 19/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0318 - val_loss: 6.7483e-04 - val_mae: 0.0266\nEpoch 20/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0013 - mae: 0.0358 - val_loss: 6.4755e-04 - val_mae: 0.0251\nEpoch 21/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0333 - val_loss: 8.1450e-04 - val_mae: 0.0307\nEpoch 22/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0010 - mae: 0.0317 - val_loss: 5.8635e-04 - val_mae: 0.0237\nEpoch 23/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.3365e-04 - mae: 0.0303 - val_loss: 9.8944e-04 - val_mae: 0.0355\nEpoch 24/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0010 - mae: 0.0315 - val_loss: 6.1671e-04 - val_mae: 0.0245\nEpoch 25/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0325 - val_loss: 6.7275e-04 - val_mae: 0.0257\nEpoch 26/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0327 - val_loss: 6.0384e-04 - val_mae: 0.0247\nEpoch 27/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0336 - val_loss: 6.2686e-04 - val_mae: 0.0249\nEpoch 28/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0010 - mae: 0.0320 - val_loss: 0.0010 - val_mae: 0.0358\nEpoch 29/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0328 - val_loss: 6.4109e-04 - val_mae: 0.0249\nEpoch 30/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0330 - val_loss: 5.8311e-04 - val_mae: 0.0239\nEpoch 31/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0316 - val_loss: 7.5583e-04 - val_mae: 0.0283\nEpoch 32/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0343 - val_loss: 0.0013 - val_mae: 0.0420\nEpoch 33/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0013 - mae: 0.0394 - val_loss: 9.1836e-04 - val_mae: 0.0341\nEpoch 34/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0327 - val_loss: 7.4189e-04 - val_mae: 0.0281\nEpoch 35/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0321 - val_loss: 7.1357e-04 - val_mae: 0.0266\nEpoch 36/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 9.8926e-04 - mae: 0.0314 - val_loss: 6.7998e-04 - val_mae: 0.0265\nEpoch 37/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 8.7518e-04 - mae: 0.0284 - val_loss: 7.2442e-04 - val_mae: 0.0281\nEpoch 38/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0336 - val_loss: 7.7866e-04 - val_mae: 0.0304\nEpoch 39/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0012 - mae: 0.0350 - val_loss: 6.1052e-04 - val_mae: 0.0243\nEpoch 40/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 9.0051e-04 - mae: 0.0288 - val_loss: 9.6522e-04 - val_mae: 0.0353\nEpoch 41/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0322 - val_loss: 0.0011 - val_mae: 0.0378\nEpoch 42/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0012 - mae: 0.0350 - val_loss: 7.3059e-04 - val_mae: 0.0269\nEpoch 43/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0011 - mae: 0.0327 - val_loss: 8.9494e-04 - val_mae: 0.0334\nEpoch 44/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0011 - mae: 0.0322 - val_loss: 6.1441e-04 - val_mae: 0.0246\nEpoch 45/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 9.7530e-04 - mae: 0.0318 - val_loss: 9.4580e-04 - val_mae: 0.0329\nEpoch 46/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 9.8872e-04 - mae: 0.0321 - val_loss: 7.5650e-04 - val_mae: 0.0271\nEpoch 47/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 9.6521e-04 - mae: 0.0306 - val_loss: 6.9529e-04 - val_mae: 0.0275\nEpoch 48/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 9.9037e-04 - mae: 0.0304 - val_loss: 6.3497e-04 - val_mae: 0.0243\nEpoch 49/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 9.5904e-04 - mae: 0.0301 - val_loss: 5.9456e-04 - val_mae: 0.0235\nEpoch 50/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 8.5539e-04 - mae: 0.0285 - val_loss: 6.3872e-04 - val_mae: 0.0245\nEpoch 51/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 8.5712e-04 - mae: 0.0283 - val_loss: 6.9115e-04 - val_mae: 0.0264\nEpoch 52/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 9.4566e-04 - mae: 0.0302 - val_loss: 6.3340e-04 - val_mae: 0.0244\nEpoch 53/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 9.4600e-04 - mae: 0.0303 - val_loss: 0.0014 - val_mae: 0.0436\nEpoch 54/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.2960e-04 - mae: 0.0313 - val_loss: 9.6369e-04 - val_mae: 0.0336\nEpoch 55/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0014 - mae: 0.0400 - val_loss: 6.7951e-04 - val_mae: 0.0254\nEpoch 56/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 8.8124e-04 - mae: 0.0290 - val_loss: 0.0010 - val_mae: 0.0365\nEpoch 57/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0326 - val_loss: 6.3321e-04 - val_mae: 0.0242\nEpoch 58/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 8.9005e-04 - mae: 0.0287 - val_loss: 0.0013 - val_mae: 0.0443\nEpoch 59/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0013 - mae: 0.0365 - val_loss: 5.8924e-04 - val_mae: 0.0231\nEpoch 60/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 8.6954e-04 - mae: 0.0280 - val_loss: 0.0013 - val_mae: 0.0439\nEpoch 61/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0341 - val_loss: 6.3933e-04 - val_mae: 0.0247\nEpoch 62/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 8.8305e-04 - mae: 0.0282 - val_loss: 6.3104e-04 - val_mae: 0.0240\nEpoch 63/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 8.6127e-04 - mae: 0.0282 - val_loss: 9.6100e-04 - val_mae: 0.0346\nEpoch 64/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0010 - mae: 0.0333 - val_loss: 6.7381e-04 - val_mae: 0.0250\nEpoch 65/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.3085e-04 - mae: 0.0299 - val_loss: 8.2203e-04 - val_mae: 0.0308\nEpoch 66/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0315 - val_loss: 6.8462e-04 - val_mae: 0.0255\nEpoch 67/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 8.7900e-04 - mae: 0.0275 - val_loss: 7.2269e-04 - val_mae: 0.0287\nEpoch 68/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0327 - val_loss: 5.4747e-04 - val_mae: 0.0227\nEpoch 69/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 9.1426e-04 - mae: 0.0300 - val_loss: 5.6545e-04 - val_mae: 0.0235\nEpoch 70/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 7.6129e-04 - mae: 0.0268 - val_loss: 6.8441e-04 - val_mae: 0.0258\nEpoch 71/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 8.0704e-04 - mae: 0.0277 - val_loss: 9.1459e-04 - val_mae: 0.0342\nEpoch 72/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 9.1146e-04 - mae: 0.0292 - val_loss: 7.2215e-04 - val_mae: 0.0278\nEpoch 73/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.1143e-04 - mae: 0.0302 - val_loss: 7.8278e-04 - val_mae: 0.0276\nEpoch 74/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0309 - val_loss: 7.9095e-04 - val_mae: 0.0300\nEpoch 75/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.4604e-04 - mae: 0.0307 - val_loss: 9.5308e-04 - val_mae: 0.0348\nEpoch 76/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0010 - mae: 0.0314 - val_loss: 6.6950e-04 - val_mae: 0.0255\nEpoch 77/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.6197e-04 - mae: 0.0298 - val_loss: 6.9510e-04 - val_mae: 0.0258\nEpoch 78/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 8.3647e-04 - mae: 0.0285 - val_loss: 6.9671e-04 - val_mae: 0.0273\nEpoch 79/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.0630e-04 - mae: 0.0290 - val_loss: 9.6537e-04 - val_mae: 0.0338\nEpoch 80/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 9.5891e-04 - mae: 0.0319 - val_loss: 9.2068e-04 - val_mae: 0.0319\nEpoch 81/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0010 - mae: 0.0320 - val_loss: 7.3118e-04 - val_mae: 0.0266\nEpoch 82/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 9.6518e-04 - mae: 0.0319 - val_loss: 6.3931e-04 - val_mae: 0.0248\nEpoch 83/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 9.0069e-04 - mae: 0.0291 - val_loss: 6.1358e-04 - val_mae: 0.0243\nEpoch 84/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 9.6971e-04 - mae: 0.0309 - val_loss: 8.0485e-04 - val_mae: 0.0278\nEpoch 85/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 9.5314e-04 - mae: 0.0301 - val_loss: 6.7118e-04 - val_mae: 0.0259\nEpoch 86/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 8.5359e-04 - mae: 0.0284 - val_loss: 6.1996e-04 - val_mae: 0.0237\nEpoch 87/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 7.9212e-04 - mae: 0.0266 - val_loss: 6.4729e-04 - val_mae: 0.0246\nEpoch 88/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 8.7475e-04 - mae: 0.0283 - val_loss: 9.2250e-04 - val_mae: 0.0338\nEpoch 89/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 8.4386e-04 - mae: 0.0291 - val_loss: 6.8795e-04 - val_mae: 0.0257\nEpoch 90/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 8.7239e-04 - mae: 0.0302 - val_loss: 6.2436e-04 - val_mae: 0.0246\nEpoch 91/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 8.6726e-04 - mae: 0.0282 - val_loss: 6.8414e-04 - val_mae: 0.0264\nEpoch 92/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.0554e-04 - mae: 0.0281 - val_loss: 7.5150e-04 - val_mae: 0.0277\nEpoch 93/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 7.3291e-04 - mae: 0.0267 - val_loss: 9.6011e-04 - val_mae: 0.0334\nEpoch 94/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 9.6917e-04 - mae: 0.0318 - val_loss: 6.8487e-04 - val_mae: 0.0262\nEpoch 95/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 8.9935e-04 - mae: 0.0298 - val_loss: 6.8605e-04 - val_mae: 0.0267\nEpoch 96/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.3017e-04 - mae: 0.0303 - val_loss: 7.5193e-04 - val_mae: 0.0275\nEpoch 97/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 8.3442e-04 - mae: 0.0292 - val_loss: 0.0011 - val_mae: 0.0363\nEpoch 98/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0328 - val_loss: 0.0015 - val_mae: 0.0477\nEpoch 99/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0325 - val_loss: 7.0543e-04 - val_mae: 0.0261\nEpoch 100/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 8.6321e-04 - mae: 0.0292 - val_loss: 6.4231e-04 - val_mae: 0.0250\nEpoch 101/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 8.4651e-04 - mae: 0.0290 - val_loss: 8.2219e-04 - val_mae: 0.0293\nEpoch 102/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 9.9352e-04 - mae: 0.0324 - val_loss: 6.5697e-04 - val_mae: 0.0250\nEpoch 103/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 8.4537e-04 - mae: 0.0276 - val_loss: 7.9343e-04 - val_mae: 0.0285\nEpoch 104/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 8.5118e-04 - mae: 0.0284 - val_loss: 9.4228e-04 - val_mae: 0.0330\nEpoch 105/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 8.6256e-04 - mae: 0.0299 - val_loss: 7.2907e-04 - val_mae: 0.0278\nEpoch 106/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0333 - val_loss: 6.5581e-04 - val_mae: 0.0253\nEpoch 107/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 7.9672e-04 - mae: 0.0275 - val_loss: 7.6418e-04 - val_mae: 0.0273\nEpoch 108/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 8.2822e-04 - mae: 0.0275 - val_loss: 6.1006e-04 - val_mae: 0.0235\nEpoch 109/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 8.7208e-04 - mae: 0.0286 - val_loss: 6.4447e-04 - val_mae: 0.0246\nEpoch 110/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 8.4023e-04 - mae: 0.0280 - val_loss: 6.8928e-04 - val_mae: 0.0259\nEpoch 111/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 8.5527e-04 - mae: 0.0290 - val_loss: 6.5470e-04 - val_mae: 0.0263\nEpoch 112/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 8.5257e-04 - mae: 0.0287 - val_loss: 6.1009e-04 - val_mae: 0.0244\nEpoch 113/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 9.9878e-04 - mae: 0.0326 - val_loss: 6.2049e-04 - val_mae: 0.0247\nEpoch 114/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 9.8441e-04 - mae: 0.0322 - val_loss: 0.0011 - val_mae: 0.0341\nEpoch 115/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 8.8528e-04 - mae: 0.0293 - val_loss: 6.4133e-04 - val_mae: 0.0244\nEpoch 116/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 9.0886e-04 - mae: 0.0291 - val_loss: 6.9066e-04 - val_mae: 0.0255\nEpoch 117/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 8.0695e-04 - mae: 0.0269 - val_loss: 7.0836e-04 - val_mae: 0.0262\nEpoch 118/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 7.7059e-04 - mae: 0.0271 - val_loss: 7.4235e-04 - val_mae: 0.0274\n\u001b[1m3/3\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 10ms/step - loss: 6.0875e-04 - mae: 0.0228\n","output_type":"stream"},{"execution_count":53,"output_type":"execute_result","data":{"text/plain":"22747.168317437172"},"metadata":{}}]},{"cell_type":"markdown","source":"## Multitask RNN that forecasts both bus and rail","metadata":{}},{"cell_type":"code","source":"# extra code – build and train a multitask RNN that forecasts both bus and rail\n\ntf.random.set_seed(42)\n\nseq_length = 56\ntrain_multask_ds = tf.keras.utils.timeseries_dataset_from_array(\n    mulvar_train.to_numpy(),\n    targets=mulvar_train[[\"bus\", \"rail\"]][seq_length:],  # 2 targets per day\n    sequence_length=seq_length,\n    batch_size=32,\n    shuffle=True,\n    seed=42\n)\nvalid_multask_ds = tf.keras.utils.timeseries_dataset_from_array(\n    mulvar_valid.to_numpy(),\n    targets=mulvar_valid[[\"bus\", \"rail\"]][seq_length:],\n    sequence_length=seq_length,\n    batch_size=32\n)\n\ntf.random.set_seed(42)\nmultask_model = tf.keras.Sequential([\n    tf.keras.layers.SimpleRNN(32, input_shape=[None, 5]),\n    tf.keras.layers.Dense(2)\n])\n\nfit_and_evaluate(multask_model, train_multask_ds, valid_multask_ds,\n                 learning_rate=0.02)","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:24:14.542918Z","iopub.execute_input":"2024-07-11T18:24:14.543691Z","iopub.status.idle":"2024-07-11T18:26:40.374318Z","shell.execute_reply.started":"2024-07-11T18:24:14.543637Z","shell.execute_reply":"2024-07-11T18:26:40.372054Z"},"trusted":true},"execution_count":54,"outputs":[{"name":"stdout","text":"Epoch 1/500\n","output_type":"stream"},{"name":"stderr","text":"/opt/conda/lib/python3.10/site-packages/keras/src/layers/rnn/rnn.py:204: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n  super().__init__(**kwargs)\n","output_type":"stream"},{"name":"stdout","text":"\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 30ms/step - loss: 0.1277 - mae: 0.3977 - val_loss: 0.0177 - val_mae: 0.1668\nEpoch 2/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0072 - mae: 0.0908 - val_loss: 0.0018 - val_mae: 0.0453\nEpoch 3/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0029 - mae: 0.0550 - val_loss: 0.0020 - val_mae: 0.0512\nEpoch 4/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0026 - mae: 0.0505 - val_loss: 0.0012 - val_mae: 0.0368\nEpoch 5/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0025 - mae: 0.0508 - val_loss: 0.0027 - val_mae: 0.0630\nEpoch 6/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0024 - mae: 0.0511 - val_loss: 0.0011 - val_mae: 0.0362\nEpoch 7/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0023 - mae: 0.0484 - val_loss: 0.0012 - val_mae: 0.0373\nEpoch 8/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0023 - mae: 0.0486 - val_loss: 9.7762e-04 - val_mae: 0.0330\nEpoch 9/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0021 - mae: 0.0473 - val_loss: 0.0014 - val_mae: 0.0412\nEpoch 10/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0021 - mae: 0.0460 - val_loss: 0.0015 - val_mae: 0.0436\nEpoch 11/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0021 - mae: 0.0453 - val_loss: 0.0011 - val_mae: 0.0360\nEpoch 12/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0020 - mae: 0.0453 - val_loss: 9.4742e-04 - val_mae: 0.0329\nEpoch 13/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0020 - mae: 0.0452 - val_loss: 0.0011 - val_mae: 0.0366\nEpoch 14/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0019 - mae: 0.0438 - val_loss: 0.0010 - val_mae: 0.0349\nEpoch 15/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0018 - mae: 0.0421 - val_loss: 0.0012 - val_mae: 0.0382\nEpoch 16/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0019 - mae: 0.0430 - val_loss: 0.0010 - val_mae: 0.0346\nEpoch 17/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0016 - mae: 0.0407 - val_loss: 8.6090e-04 - val_mae: 0.0322\nEpoch 18/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0020 - mae: 0.0444 - val_loss: 0.0015 - val_mae: 0.0454\nEpoch 19/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0018 - mae: 0.0423 - val_loss: 0.0011 - val_mae: 0.0368\nEpoch 20/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0018 - mae: 0.0416 - val_loss: 9.3085e-04 - val_mae: 0.0327\nEpoch 21/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0018 - mae: 0.0415 - val_loss: 8.7311e-04 - val_mae: 0.0325\nEpoch 22/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0016 - mae: 0.0416 - val_loss: 9.9171e-04 - val_mae: 0.0343\nEpoch 23/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0015 - mae: 0.0393 - val_loss: 0.0013 - val_mae: 0.0404\nEpoch 24/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0016 - mae: 0.0402 - val_loss: 0.0011 - val_mae: 0.0358\nEpoch 25/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0016 - mae: 0.0406 - val_loss: 9.5088e-04 - val_mae: 0.0334\nEpoch 26/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0016 - mae: 0.0404 - val_loss: 9.7570e-04 - val_mae: 0.0339\nEpoch 27/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0016 - mae: 0.0404 - val_loss: 0.0012 - val_mae: 0.0394\nEpoch 28/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0015 - mae: 0.0381 - val_loss: 8.5759e-04 - val_mae: 0.0320\nEpoch 29/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0016 - mae: 0.0399 - val_loss: 7.8438e-04 - val_mae: 0.0298\nEpoch 30/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0016 - mae: 0.0397 - val_loss: 8.2246e-04 - val_mae: 0.0303\nEpoch 31/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0016 - mae: 0.0388 - val_loss: 9.1792e-04 - val_mae: 0.0325\nEpoch 32/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0015 - mae: 0.0386 - val_loss: 0.0011 - val_mae: 0.0366\nEpoch 33/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0014 - mae: 0.0378 - val_loss: 7.4688e-04 - val_mae: 0.0296\nEpoch 34/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0016 - mae: 0.0402 - val_loss: 8.4005e-04 - val_mae: 0.0308\nEpoch 35/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0015 - mae: 0.0375 - val_loss: 9.6971e-04 - val_mae: 0.0336\nEpoch 36/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0014 - mae: 0.0371 - val_loss: 0.0010 - val_mae: 0.0356\nEpoch 37/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0013 - mae: 0.0368 - val_loss: 8.5215e-04 - val_mae: 0.0322\nEpoch 38/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0017 - mae: 0.0402 - val_loss: 7.0198e-04 - val_mae: 0.0272\nEpoch 39/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0014 - mae: 0.0369 - val_loss: 7.6073e-04 - val_mae: 0.0287\nEpoch 40/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0014 - mae: 0.0376 - val_loss: 9.2714e-04 - val_mae: 0.0331\nEpoch 41/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0015 - mae: 0.0372 - val_loss: 9.8305e-04 - val_mae: 0.0343\nEpoch 42/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0013 - mae: 0.0359 - val_loss: 8.7180e-04 - val_mae: 0.0317\nEpoch 43/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0014 - mae: 0.0372 - val_loss: 9.7888e-04 - val_mae: 0.0343\nEpoch 44/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0014 - mae: 0.0368 - val_loss: 9.3103e-04 - val_mae: 0.0330\nEpoch 45/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0013 - mae: 0.0362 - val_loss: 8.7919e-04 - val_mae: 0.0317\nEpoch 46/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0012 - mae: 0.0350 - val_loss: 8.6372e-04 - val_mae: 0.0314\nEpoch 47/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0014 - mae: 0.0370 - val_loss: 8.2071e-04 - val_mae: 0.0303\nEpoch 48/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0015 - mae: 0.0373 - val_loss: 7.8473e-04 - val_mae: 0.0294\nEpoch 49/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0014 - mae: 0.0355 - val_loss: 7.2143e-04 - val_mae: 0.0280\nEpoch 50/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0349 - val_loss: 8.5807e-04 - val_mae: 0.0310\nEpoch 51/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0342 - val_loss: 0.0011 - val_mae: 0.0373\nEpoch 52/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0013 - mae: 0.0356 - val_loss: 8.4246e-04 - val_mae: 0.0318\nEpoch 53/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0013 - mae: 0.0359 - val_loss: 9.0097e-04 - val_mae: 0.0322\nEpoch 54/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0012 - mae: 0.0345 - val_loss: 8.1703e-04 - val_mae: 0.0301\nEpoch 55/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0014 - mae: 0.0383 - val_loss: 7.1553e-04 - val_mae: 0.0276\nEpoch 56/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0013 - mae: 0.0349 - val_loss: 8.3689e-04 - val_mae: 0.0307\nEpoch 57/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0013 - mae: 0.0353 - val_loss: 7.9783e-04 - val_mae: 0.0298\nEpoch 58/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0344 - val_loss: 7.6952e-04 - val_mae: 0.0291\nEpoch 59/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0013 - mae: 0.0357 - val_loss: 7.1594e-04 - val_mae: 0.0275\nEpoch 60/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0012 - mae: 0.0338 - val_loss: 7.7117e-04 - val_mae: 0.0292\nEpoch 61/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0344 - val_loss: 7.7873e-04 - val_mae: 0.0295\nEpoch 62/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0013 - mae: 0.0350 - val_loss: 7.7749e-04 - val_mae: 0.0292\nEpoch 63/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0339 - val_loss: 7.5299e-04 - val_mae: 0.0286\nEpoch 64/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0347 - val_loss: 7.1668e-04 - val_mae: 0.0283\nEpoch 65/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0012 - mae: 0.0346 - val_loss: 8.5140e-04 - val_mae: 0.0314\nEpoch 66/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0014 - mae: 0.0359 - val_loss: 6.9599e-04 - val_mae: 0.0277\nEpoch 67/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0340 - val_loss: 8.4997e-04 - val_mae: 0.0310\nEpoch 68/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0014 - mae: 0.0358 - val_loss: 6.8885e-04 - val_mae: 0.0279\nEpoch 69/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0012 - mae: 0.0340 - val_loss: 6.8942e-04 - val_mae: 0.0277\nEpoch 70/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0010 - mae: 0.0320 - val_loss: 7.2840e-04 - val_mae: 0.0288\nEpoch 71/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0012 - mae: 0.0335 - val_loss: 6.9064e-04 - val_mae: 0.0279\nEpoch 72/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0337 - val_loss: 6.6668e-04 - val_mae: 0.0267\nEpoch 73/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0011 - mae: 0.0329 - val_loss: 9.8947e-04 - val_mae: 0.0342\nEpoch 74/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0013 - mae: 0.0354 - val_loss: 6.6038e-04 - val_mae: 0.0265\nEpoch 75/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0012 - mae: 0.0335 - val_loss: 6.6545e-04 - val_mae: 0.0269\nEpoch 76/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0012 - mae: 0.0339 - val_loss: 7.6584e-04 - val_mae: 0.0289\nEpoch 77/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0013 - mae: 0.0350 - val_loss: 6.8700e-04 - val_mae: 0.0266\nEpoch 78/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0011 - mae: 0.0317 - val_loss: 6.8237e-04 - val_mae: 0.0267\nEpoch 79/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0011 - mae: 0.0323 - val_loss: 6.8346e-04 - val_mae: 0.0277\nEpoch 80/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0012 - mae: 0.0349 - val_loss: 6.9324e-04 - val_mae: 0.0280\nEpoch 81/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0331 - val_loss: 7.7558e-04 - val_mae: 0.0292\nEpoch 82/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0012 - mae: 0.0340 - val_loss: 7.7804e-04 - val_mae: 0.0292\nEpoch 83/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0013 - mae: 0.0340 - val_loss: 6.6594e-04 - val_mae: 0.0265\nEpoch 84/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0012 - mae: 0.0335 - val_loss: 7.0078e-04 - val_mae: 0.0282\nEpoch 85/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0012 - mae: 0.0337 - val_loss: 6.5370e-04 - val_mae: 0.0258\nEpoch 86/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0011 - mae: 0.0330 - val_loss: 7.1686e-04 - val_mae: 0.0287\nEpoch 87/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0011 - mae: 0.0335 - val_loss: 6.6447e-04 - val_mae: 0.0269\nEpoch 88/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0012 - mae: 0.0334 - val_loss: 6.7835e-04 - val_mae: 0.0269\nEpoch 89/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0010 - mae: 0.0319 - val_loss: 6.4167e-04 - val_mae: 0.0254\nEpoch 90/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0013 - mae: 0.0363 - val_loss: 7.5077e-04 - val_mae: 0.0285\nEpoch 91/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0012 - mae: 0.0325 - val_loss: 7.1167e-04 - val_mae: 0.0281\nEpoch 92/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0013 - mae: 0.0342 - val_loss: 6.7780e-04 - val_mae: 0.0267\nEpoch 93/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0342 - val_loss: 6.8547e-04 - val_mae: 0.0275\nEpoch 94/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0013 - mae: 0.0355 - val_loss: 6.3924e-04 - val_mae: 0.0253\nEpoch 95/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0342 - val_loss: 0.0011 - val_mae: 0.0374\nEpoch 96/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0348 - val_loss: 8.5004e-04 - val_mae: 0.0319\nEpoch 97/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0355 - val_loss: 6.5320e-04 - val_mae: 0.0262\nEpoch 98/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0012 - mae: 0.0336 - val_loss: 8.4479e-04 - val_mae: 0.0309\nEpoch 99/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0326 - val_loss: 8.4633e-04 - val_mae: 0.0313\nEpoch 100/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0012 - mae: 0.0342 - val_loss: 7.8719e-04 - val_mae: 0.0299\nEpoch 101/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0340 - val_loss: 8.2049e-04 - val_mae: 0.0304\nEpoch 102/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0326 - val_loss: 7.9130e-04 - val_mae: 0.0296\nEpoch 103/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0330 - val_loss: 6.4601e-04 - val_mae: 0.0257\nEpoch 104/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0325 - val_loss: 6.3770e-04 - val_mae: 0.0256\nEpoch 105/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0332 - val_loss: 6.8253e-04 - val_mae: 0.0275\nEpoch 106/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0346 - val_loss: 7.2086e-04 - val_mae: 0.0275\nEpoch 107/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0012 - mae: 0.0335 - val_loss: 6.5663e-04 - val_mae: 0.0264\nEpoch 108/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0012 - mae: 0.0329 - val_loss: 6.6241e-04 - val_mae: 0.0262\nEpoch 109/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0330 - val_loss: 6.5807e-04 - val_mae: 0.0264\nEpoch 110/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0330 - val_loss: 8.1531e-04 - val_mae: 0.0303\nEpoch 111/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0330 - val_loss: 7.1091e-04 - val_mae: 0.0275\nEpoch 112/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0324 - val_loss: 6.8647e-04 - val_mae: 0.0274\nEpoch 113/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0325 - val_loss: 6.3750e-04 - val_mae: 0.0254\nEpoch 114/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.8613e-04 - mae: 0.0311 - val_loss: 7.0207e-04 - val_mae: 0.0276\nEpoch 115/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0324 - val_loss: 7.2694e-04 - val_mae: 0.0281\nEpoch 116/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0012 - mae: 0.0333 - val_loss: 6.3905e-04 - val_mae: 0.0254\nEpoch 117/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0011 - mae: 0.0326 - val_loss: 6.4325e-04 - val_mae: 0.0259\nEpoch 118/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0011 - mae: 0.0314 - val_loss: 7.2266e-04 - val_mae: 0.0275\nEpoch 119/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0011 - mae: 0.0332 - val_loss: 6.7352e-04 - val_mae: 0.0265\nEpoch 120/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0013 - mae: 0.0353 - val_loss: 7.0674e-04 - val_mae: 0.0280\nEpoch 121/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0012 - mae: 0.0341 - val_loss: 9.6777e-04 - val_mae: 0.0342\nEpoch 122/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0013 - mae: 0.0345 - val_loss: 6.4701e-04 - val_mae: 0.0257\nEpoch 123/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0011 - mae: 0.0321 - val_loss: 7.1019e-04 - val_mae: 0.0275\nEpoch 124/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0012 - mae: 0.0334 - val_loss: 6.4270e-04 - val_mae: 0.0258\nEpoch 125/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0321 - val_loss: 6.6076e-04 - val_mae: 0.0261\nEpoch 126/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0011 - mae: 0.0330 - val_loss: 6.4890e-04 - val_mae: 0.0257\nEpoch 127/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0317 - val_loss: 6.6202e-04 - val_mae: 0.0268\nEpoch 128/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0319 - val_loss: 8.7491e-04 - val_mae: 0.0318\nEpoch 129/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0332 - val_loss: 7.8031e-04 - val_mae: 0.0294\nEpoch 130/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0332 - val_loss: 8.3239e-04 - val_mae: 0.0308\nEpoch 131/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0325 - val_loss: 6.4946e-04 - val_mae: 0.0264\nEpoch 132/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0010 - mae: 0.0310 - val_loss: 6.7604e-04 - val_mae: 0.0269\nEpoch 133/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0320 - val_loss: 6.5455e-04 - val_mae: 0.0268\nEpoch 134/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0317 - val_loss: 7.3712e-04 - val_mae: 0.0285\nEpoch 135/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0321 - val_loss: 7.1683e-04 - val_mae: 0.0277\nEpoch 136/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0326 - val_loss: 6.4048e-04 - val_mae: 0.0259\nEpoch 137/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0010 - mae: 0.0313 - val_loss: 6.9906e-04 - val_mae: 0.0277\nEpoch 138/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0328 - val_loss: 6.4916e-04 - val_mae: 0.0260\nEpoch 139/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.7453e-04 - mae: 0.0309 - val_loss: 7.4788e-04 - val_mae: 0.0285\nEpoch 140/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0324 - val_loss: 6.3452e-04 - val_mae: 0.0251\nEpoch 141/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0010 - mae: 0.0314 - val_loss: 6.6216e-04 - val_mae: 0.0267\nEpoch 142/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0333 - val_loss: 7.7436e-04 - val_mae: 0.0294\nEpoch 143/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0321 - val_loss: 6.5625e-04 - val_mae: 0.0261\nEpoch 144/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0321 - val_loss: 6.7637e-04 - val_mae: 0.0264\nEpoch 145/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 9.8056e-04 - mae: 0.0312 - val_loss: 6.5679e-04 - val_mae: 0.0254\nEpoch 146/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0307 - val_loss: 6.9308e-04 - val_mae: 0.0270\nEpoch 147/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0321 - val_loss: 9.1986e-04 - val_mae: 0.0333\nEpoch 148/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0013 - mae: 0.0351 - val_loss: 6.4982e-04 - val_mae: 0.0260\nEpoch 149/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.5500e-04 - mae: 0.0309 - val_loss: 7.2336e-04 - val_mae: 0.0281\nEpoch 150/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0010 - mae: 0.0315 - val_loss: 7.8625e-04 - val_mae: 0.0297\nEpoch 151/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0326 - val_loss: 6.6486e-04 - val_mae: 0.0265\nEpoch 152/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0010 - mae: 0.0311 - val_loss: 6.8714e-04 - val_mae: 0.0263\nEpoch 153/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0313 - val_loss: 7.0510e-04 - val_mae: 0.0279\nEpoch 154/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0320 - val_loss: 6.2500e-04 - val_mae: 0.0251\nEpoch 155/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0308 - val_loss: 6.3489e-04 - val_mae: 0.0248\nEpoch 156/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 9.8001e-04 - mae: 0.0314 - val_loss: 6.4826e-04 - val_mae: 0.0253\nEpoch 157/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0308 - val_loss: 7.0555e-04 - val_mae: 0.0274\nEpoch 158/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0010 - mae: 0.0312 - val_loss: 9.4039e-04 - val_mae: 0.0338\nEpoch 159/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0329 - val_loss: 7.1961e-04 - val_mae: 0.0276\nEpoch 160/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0011 - mae: 0.0316 - val_loss: 6.4454e-04 - val_mae: 0.0260\nEpoch 161/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 9.2960e-04 - mae: 0.0301 - val_loss: 7.4932e-04 - val_mae: 0.0286\nEpoch 162/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0317 - val_loss: 6.6614e-04 - val_mae: 0.0262\nEpoch 163/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0011 - mae: 0.0316 - val_loss: 6.4117e-04 - val_mae: 0.0259\nEpoch 164/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0010 - mae: 0.0311 - val_loss: 6.8727e-04 - val_mae: 0.0271\nEpoch 165/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0011 - mae: 0.0322 - val_loss: 6.3960e-04 - val_mae: 0.0258\nEpoch 166/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0011 - mae: 0.0326 - val_loss: 7.1289e-04 - val_mae: 0.0275\nEpoch 167/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0011 - mae: 0.0326 - val_loss: 6.6839e-04 - val_mae: 0.0261\nEpoch 168/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 9.7505e-04 - mae: 0.0305 - val_loss: 8.4228e-04 - val_mae: 0.0313\nEpoch 169/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0012 - mae: 0.0338 - val_loss: 6.3355e-04 - val_mae: 0.0249\nEpoch 170/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.3815e-04 - mae: 0.0299 - val_loss: 6.5543e-04 - val_mae: 0.0260\nEpoch 171/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0010 - mae: 0.0308 - val_loss: 6.5528e-04 - val_mae: 0.0258\nEpoch 172/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0310 - val_loss: 6.3014e-04 - val_mae: 0.0252\nEpoch 173/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0322 - val_loss: 7.4493e-04 - val_mae: 0.0282\nEpoch 174/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0325 - val_loss: 6.8670e-04 - val_mae: 0.0271\nEpoch 175/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 9.8753e-04 - mae: 0.0314 - val_loss: 7.2202e-04 - val_mae: 0.0278\nEpoch 176/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0012 - mae: 0.0332 - val_loss: 6.5481e-04 - val_mae: 0.0259\nEpoch 177/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0320 - val_loss: 6.4482e-04 - val_mae: 0.0258\nEpoch 178/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0308 - val_loss: 6.5569e-04 - val_mae: 0.0261\nEpoch 179/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.7566e-04 - mae: 0.0308 - val_loss: 6.5018e-04 - val_mae: 0.0262\nEpoch 180/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.9544e-04 - mae: 0.0306 - val_loss: 6.7548e-04 - val_mae: 0.0264\nEpoch 181/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 9.5005e-04 - mae: 0.0305 - val_loss: 7.5716e-04 - val_mae: 0.0293\nEpoch 182/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0010 - mae: 0.0312 - val_loss: 7.2054e-04 - val_mae: 0.0277\nEpoch 183/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0316 - val_loss: 6.4615e-04 - val_mae: 0.0253\nEpoch 184/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0321 - val_loss: 6.2950e-04 - val_mae: 0.0252\nEpoch 185/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0011 - mae: 0.0313 - val_loss: 6.7657e-04 - val_mae: 0.0268\nEpoch 186/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.9679e-04 - mae: 0.0311 - val_loss: 6.5428e-04 - val_mae: 0.0255\nEpoch 187/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0312 - val_loss: 7.5176e-04 - val_mae: 0.0287\nEpoch 188/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0318 - val_loss: 6.9797e-04 - val_mae: 0.0273\nEpoch 189/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0314 - val_loss: 8.4907e-04 - val_mae: 0.0315\nEpoch 190/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0324 - val_loss: 6.3531e-04 - val_mae: 0.0254\nEpoch 191/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 9.8856e-04 - mae: 0.0310 - val_loss: 6.8022e-04 - val_mae: 0.0266\nEpoch 192/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0310 - val_loss: 6.6849e-04 - val_mae: 0.0261\nEpoch 193/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.2544e-04 - mae: 0.0297 - val_loss: 6.6309e-04 - val_mae: 0.0261\nEpoch 194/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.2524e-04 - mae: 0.0301 - val_loss: 6.4715e-04 - val_mae: 0.0260\nEpoch 195/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0315 - val_loss: 7.0957e-04 - val_mae: 0.0275\nEpoch 196/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.9767e-04 - mae: 0.0307 - val_loss: 7.0033e-04 - val_mae: 0.0268\nEpoch 197/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0010 - mae: 0.0316 - val_loss: 7.1106e-04 - val_mae: 0.0275\nEpoch 198/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0011 - mae: 0.0326 - val_loss: 6.8635e-04 - val_mae: 0.0265\nEpoch 199/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 9.6868e-04 - mae: 0.0301 - val_loss: 6.9426e-04 - val_mae: 0.0269\nEpoch 200/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0010 - mae: 0.0316 - val_loss: 6.4064e-04 - val_mae: 0.0254\nEpoch 201/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0010 - mae: 0.0307 - val_loss: 8.5646e-04 - val_mae: 0.0317\nEpoch 202/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0011 - mae: 0.0333 - val_loss: 6.5845e-04 - val_mae: 0.0263\nEpoch 203/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0010 - mae: 0.0310 - val_loss: 8.3874e-04 - val_mae: 0.0310\nEpoch 204/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0011 - mae: 0.0330 - val_loss: 6.6275e-04 - val_mae: 0.0266\nEpoch 205/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0012 - mae: 0.0324 - val_loss: 6.4835e-04 - val_mae: 0.0259\n\u001b[1m3/3\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 10ms/step - loss: 6.3559e-04 - mae: 0.0240\n","output_type":"stream"},{"execution_count":54,"output_type":"execute_result","data":{"text/plain":"24803.616106510162"},"metadata":{}}]},{"cell_type":"markdown","source":"# Forecasting Several Time Steps Ahead","metadata":{}},{"cell_type":"code","source":"# we could just change the targets to be the value 14 days ahead instead of 1 day ahead.\n\nX = rail_valid.to_numpy()[np.newaxis, :seq_length, np.newaxis]\nfor step_ahead in range(14):\n y_pred_one = univar_model.predict(X)\n X = np.concatenate([X, y_pred_one.reshape(1, 1, 1)], axis=1)\n","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:29:22.054881Z","iopub.execute_input":"2024-07-11T18:29:22.055489Z","iopub.status.idle":"2024-07-11T18:29:23.696262Z","shell.execute_reply.started":"2024-07-11T18:29:22.055449Z","shell.execute_reply":"2024-07-11T18:29:23.694648Z"},"trusted":true},"execution_count":55,"outputs":[{"name":"stdout","text":"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 178ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 183ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n","output_type":"stream"}]},{"cell_type":"code","source":"# The forecasts start on 2019-02-26, as it is the 57th day of 2019, and they end\n# on 2019-03-11. That's 14 days in total.\nY_pred = pd.Series(X[0, -14:, 0],\n                   index=pd.date_range(\"2019-02-26\", \"2019-03-11\"))\n\nfig, ax = plt.subplots(figsize=(8, 3.5))\n(rail_valid * 1e6)[\"2019-02-01\":\"2019-03-11\"].plot(\n    label=\"True\", marker=\".\", ax=ax)\n(Y_pred * 1e6).plot(\n    label=\"Predictions\", grid=True, marker=\"x\", color=\"r\", ax=ax)\nax.vlines(\"2019-02-25\", 0, 1e6, color=\"k\", linestyle=\"--\", label=\"Today\")\nax.set_ylim([200_000, 800_000])\nplt.legend(loc=\"center left\")\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:32:04.143757Z","iopub.execute_input":"2024-07-11T18:32:04.144265Z","iopub.status.idle":"2024-07-11T18:32:04.862663Z","shell.execute_reply.started":"2024-07-11T18:32:04.144225Z","shell.execute_reply":"2024-07-11T18:32:04.860870Z"},"trusted":true},"execution_count":57,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 800x350 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"def split_inputs_and_targets(mulvar_series, ahead=14, target_col=1):\n    return mulvar_series[:, :-ahead], mulvar_series[:, -ahead:, target_col]\n\nahead_train_ds = tf.keras.utils.timeseries_dataset_from_array(\n    mulvar_train.to_numpy(),\n    targets=None,\n    sequence_length=seq_length + 14,\n    batch_size=32,\n    shuffle=True,\n    seed=42\n).map(split_inputs_and_targets)\nahead_valid_ds = tf.keras.utils.timeseries_dataset_from_array(\n    mulvar_valid.to_numpy(),\n    targets=None,\n    sequence_length=seq_length + 14,\n    batch_size=32\n).map(split_inputs_and_targets)","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:32:36.339797Z","iopub.execute_input":"2024-07-11T18:32:36.341210Z","iopub.status.idle":"2024-07-11T18:32:36.570455Z","shell.execute_reply.started":"2024-07-11T18:32:36.341160Z","shell.execute_reply":"2024-07-11T18:32:36.568823Z"},"trusted":true},"execution_count":58,"outputs":[]},{"cell_type":"code","source":"tf.random.set_seed(42)\n\nahead_model = tf.keras.Sequential([\n    tf.keras.layers.SimpleRNN(32, input_shape=[None, 5]),\n    tf.keras.layers.Dense(14)\n])","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:32:47.195479Z","iopub.execute_input":"2024-07-11T18:32:47.196035Z","iopub.status.idle":"2024-07-11T18:32:47.295767Z","shell.execute_reply.started":"2024-07-11T18:32:47.195995Z","shell.execute_reply":"2024-07-11T18:32:47.294034Z"},"trusted":true},"execution_count":59,"outputs":[{"name":"stderr","text":"/opt/conda/lib/python3.10/site-packages/keras/src/layers/rnn/rnn.py:204: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n  super().__init__(**kwargs)\n","output_type":"stream"}]},{"cell_type":"code","source":"fit_and_evaluate(ahead_model, ahead_train_ds, ahead_valid_ds,\n                 learning_rate=0.02)","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:33:00.070886Z","iopub.execute_input":"2024-07-11T18:33:00.071380Z","iopub.status.idle":"2024-07-11T18:35:36.048833Z","shell.execute_reply.started":"2024-07-11T18:33:00.071344Z","shell.execute_reply":"2024-07-11T18:35:36.047245Z"},"trusted":true},"execution_count":60,"outputs":[{"name":"stdout","text":"Epoch 1/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 24ms/step - loss: 0.1542 - mae: 0.4333 - val_loss: 0.0207 - val_mae: 0.1552\nEpoch 2/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0182 - mae: 0.1545 - val_loss: 0.0137 - val_mae: 0.1238\nEpoch 3/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0143 - mae: 0.1343 - val_loss: 0.0120 - val_mae: 0.1174\nEpoch 4/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0127 - mae: 0.1261 - val_loss: 0.0106 - val_mae: 0.1107\nEpoch 5/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0113 - mae: 0.1182 - val_loss: 0.0093 - val_mae: 0.1038\nEpoch 6/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0106 - mae: 0.1144 - val_loss: 0.0086 - val_mae: 0.0981\nEpoch 7/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0097 - mae: 0.1069 - val_loss: 0.0074 - val_mae: 0.0921\nEpoch 8/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0088 - mae: 0.1019 - val_loss: 0.0067 - val_mae: 0.0865\nEpoch 9/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0083 - mae: 0.0979 - val_loss: 0.0058 - val_mae: 0.0807\nEpoch 10/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0074 - mae: 0.0924 - val_loss: 0.0053 - val_mae: 0.0766\nEpoch 11/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0071 - mae: 0.0890 - val_loss: 0.0047 - val_mae: 0.0721\nEpoch 12/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0067 - mae: 0.0851 - val_loss: 0.0041 - val_mae: 0.0683\nEpoch 13/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0063 - mae: 0.0821 - val_loss: 0.0037 - val_mae: 0.0644\nEpoch 14/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0061 - mae: 0.0808 - val_loss: 0.0037 - val_mae: 0.0646\nEpoch 15/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0053 - mae: 0.0738 - val_loss: 0.0033 - val_mae: 0.0619\nEpoch 16/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0054 - mae: 0.0741 - val_loss: 0.0032 - val_mae: 0.0604\nEpoch 17/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0052 - mae: 0.0715 - val_loss: 0.0030 - val_mae: 0.0585\nEpoch 18/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0053 - mae: 0.0715 - val_loss: 0.0030 - val_mae: 0.0581\nEpoch 19/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0051 - mae: 0.0697 - val_loss: 0.0027 - val_mae: 0.0556\nEpoch 20/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0050 - mae: 0.0687 - val_loss: 0.0026 - val_mae: 0.0546\nEpoch 21/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0049 - mae: 0.0673 - val_loss: 0.0027 - val_mae: 0.0560\nEpoch 22/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0049 - mae: 0.0663 - val_loss: 0.0025 - val_mae: 0.0533\nEpoch 23/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0049 - mae: 0.0664 - val_loss: 0.0025 - val_mae: 0.0539\nEpoch 24/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0046 - mae: 0.0648 - val_loss: 0.0025 - val_mae: 0.0535\nEpoch 25/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0049 - mae: 0.0662 - val_loss: 0.0025 - val_mae: 0.0536\nEpoch 26/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0046 - mae: 0.0633 - val_loss: 0.0023 - val_mae: 0.0503\nEpoch 27/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0047 - mae: 0.0647 - val_loss: 0.0023 - val_mae: 0.0501\nEpoch 28/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0047 - mae: 0.0642 - val_loss: 0.0023 - val_mae: 0.0508\nEpoch 29/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0046 - mae: 0.0625 - val_loss: 0.0022 - val_mae: 0.0490\nEpoch 30/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0044 - mae: 0.0619 - val_loss: 0.0022 - val_mae: 0.0497\nEpoch 31/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0043 - mae: 0.0609 - val_loss: 0.0021 - val_mae: 0.0476\nEpoch 32/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0047 - mae: 0.0644 - val_loss: 0.0021 - val_mae: 0.0487\nEpoch 33/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0045 - mae: 0.0619 - val_loss: 0.0020 - val_mae: 0.0469\nEpoch 34/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0042 - mae: 0.0608 - val_loss: 0.0021 - val_mae: 0.0477\nEpoch 35/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0043 - mae: 0.0604 - val_loss: 0.0020 - val_mae: 0.0469\nEpoch 36/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0044 - mae: 0.0608 - val_loss: 0.0021 - val_mae: 0.0483\nEpoch 37/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0045 - mae: 0.0613 - val_loss: 0.0021 - val_mae: 0.0477\nEpoch 38/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0041 - mae: 0.0587 - val_loss: 0.0020 - val_mae: 0.0465\nEpoch 39/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0042 - mae: 0.0595 - val_loss: 0.0021 - val_mae: 0.0488\nEpoch 40/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0044 - mae: 0.0604 - val_loss: 0.0020 - val_mae: 0.0464\nEpoch 41/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0042 - mae: 0.0587 - val_loss: 0.0020 - val_mae: 0.0466\nEpoch 42/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0042 - mae: 0.0587 - val_loss: 0.0019 - val_mae: 0.0456\nEpoch 43/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0041 - mae: 0.0581 - val_loss: 0.0020 - val_mae: 0.0463\nEpoch 44/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0042 - mae: 0.0587 - val_loss: 0.0018 - val_mae: 0.0442\nEpoch 45/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0043 - mae: 0.0592 - val_loss: 0.0019 - val_mae: 0.0452\nEpoch 46/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0042 - mae: 0.0585 - val_loss: 0.0019 - val_mae: 0.0457\nEpoch 47/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0041 - mae: 0.0580 - val_loss: 0.0019 - val_mae: 0.0451\nEpoch 48/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0042 - mae: 0.0578 - val_loss: 0.0019 - val_mae: 0.0450\nEpoch 49/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0043 - mae: 0.0581 - val_loss: 0.0017 - val_mae: 0.0425\nEpoch 50/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0043 - mae: 0.0590 - val_loss: 0.0018 - val_mae: 0.0442\nEpoch 51/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0042 - mae: 0.0579 - val_loss: 0.0018 - val_mae: 0.0431\nEpoch 52/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0040 - mae: 0.0572 - val_loss: 0.0018 - val_mae: 0.0433\nEpoch 53/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0043 - mae: 0.0587 - val_loss: 0.0018 - val_mae: 0.0431\nEpoch 54/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0040 - mae: 0.0564 - val_loss: 0.0017 - val_mae: 0.0425\nEpoch 55/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0041 - mae: 0.0577 - val_loss: 0.0017 - val_mae: 0.0420\nEpoch 56/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0041 - mae: 0.0569 - val_loss: 0.0019 - val_mae: 0.0452\nEpoch 57/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0040 - mae: 0.0564 - val_loss: 0.0017 - val_mae: 0.0426\nEpoch 58/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0040 - mae: 0.0573 - val_loss: 0.0019 - val_mae: 0.0446\nEpoch 59/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0039 - mae: 0.0559 - val_loss: 0.0018 - val_mae: 0.0430\nEpoch 60/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0042 - mae: 0.0567 - val_loss: 0.0017 - val_mae: 0.0422\nEpoch 61/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0039 - mae: 0.0559 - val_loss: 0.0016 - val_mae: 0.0408\nEpoch 62/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0039 - mae: 0.0553 - val_loss: 0.0017 - val_mae: 0.0423\nEpoch 63/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0039 - mae: 0.0553 - val_loss: 0.0018 - val_mae: 0.0438\nEpoch 64/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0040 - mae: 0.0559 - val_loss: 0.0017 - val_mae: 0.0429\nEpoch 65/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0039 - mae: 0.0552 - val_loss: 0.0017 - val_mae: 0.0428\nEpoch 66/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0041 - mae: 0.0557 - val_loss: 0.0018 - val_mae: 0.0437\nEpoch 67/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0039 - mae: 0.0549 - val_loss: 0.0015 - val_mae: 0.0388\nEpoch 68/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0039 - mae: 0.0573 - val_loss: 0.0019 - val_mae: 0.0447\nEpoch 69/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0040 - mae: 0.0559 - val_loss: 0.0018 - val_mae: 0.0431\nEpoch 70/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0039 - mae: 0.0551 - val_loss: 0.0017 - val_mae: 0.0419\nEpoch 71/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0038 - mae: 0.0545 - val_loss: 0.0017 - val_mae: 0.0415\nEpoch 72/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0039 - mae: 0.0556 - val_loss: 0.0018 - val_mae: 0.0431\nEpoch 73/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0039 - mae: 0.0547 - val_loss: 0.0016 - val_mae: 0.0405\nEpoch 74/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0038 - mae: 0.0548 - val_loss: 0.0016 - val_mae: 0.0408\nEpoch 75/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0038 - mae: 0.0540 - val_loss: 0.0018 - val_mae: 0.0433\nEpoch 76/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0038 - mae: 0.0543 - val_loss: 0.0017 - val_mae: 0.0425\nEpoch 77/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0039 - mae: 0.0543 - val_loss: 0.0018 - val_mae: 0.0432\nEpoch 78/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0037 - mae: 0.0533 - val_loss: 0.0016 - val_mae: 0.0403\nEpoch 79/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0039 - mae: 0.0550 - val_loss: 0.0017 - val_mae: 0.0425\nEpoch 80/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0039 - mae: 0.0548 - val_loss: 0.0016 - val_mae: 0.0396\nEpoch 81/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0039 - mae: 0.0544 - val_loss: 0.0016 - val_mae: 0.0404\nEpoch 82/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0038 - mae: 0.0539 - val_loss: 0.0016 - val_mae: 0.0397\nEpoch 83/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0039 - mae: 0.0548 - val_loss: 0.0016 - val_mae: 0.0401\nEpoch 84/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0039 - mae: 0.0548 - val_loss: 0.0016 - val_mae: 0.0394\nEpoch 85/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0039 - mae: 0.0547 - val_loss: 0.0017 - val_mae: 0.0421\nEpoch 86/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0039 - mae: 0.0536 - val_loss: 0.0016 - val_mae: 0.0399\nEpoch 87/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0039 - mae: 0.0544 - val_loss: 0.0015 - val_mae: 0.0387\nEpoch 88/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0039 - mae: 0.0551 - val_loss: 0.0016 - val_mae: 0.0399\nEpoch 89/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0037 - mae: 0.0527 - val_loss: 0.0016 - val_mae: 0.0397\nEpoch 90/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0037 - mae: 0.0535 - val_loss: 0.0016 - val_mae: 0.0399\nEpoch 91/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0038 - mae: 0.0538 - val_loss: 0.0016 - val_mae: 0.0391\nEpoch 92/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0039 - mae: 0.0541 - val_loss: 0.0016 - val_mae: 0.0393\nEpoch 93/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0038 - mae: 0.0537 - val_loss: 0.0017 - val_mae: 0.0410\nEpoch 94/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0037 - mae: 0.0527 - val_loss: 0.0015 - val_mae: 0.0372\nEpoch 95/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0038 - mae: 0.0544 - val_loss: 0.0015 - val_mae: 0.0382\nEpoch 96/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0037 - mae: 0.0528 - val_loss: 0.0017 - val_mae: 0.0410\nEpoch 97/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0038 - mae: 0.0533 - val_loss: 0.0015 - val_mae: 0.0386\nEpoch 98/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0036 - mae: 0.0521 - val_loss: 0.0016 - val_mae: 0.0399\nEpoch 99/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0037 - mae: 0.0523 - val_loss: 0.0015 - val_mae: 0.0377\nEpoch 100/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0038 - mae: 0.0530 - val_loss: 0.0017 - val_mae: 0.0411\nEpoch 101/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0037 - mae: 0.0526 - val_loss: 0.0015 - val_mae: 0.0376\nEpoch 102/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0035 - mae: 0.0516 - val_loss: 0.0017 - val_mae: 0.0414\nEpoch 103/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0037 - mae: 0.0526 - val_loss: 0.0016 - val_mae: 0.0392\nEpoch 104/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0038 - mae: 0.0530 - val_loss: 0.0015 - val_mae: 0.0374\nEpoch 105/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0038 - mae: 0.0537 - val_loss: 0.0014 - val_mae: 0.0365\nEpoch 106/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0036 - mae: 0.0522 - val_loss: 0.0016 - val_mae: 0.0397\nEpoch 107/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0036 - mae: 0.0520 - val_loss: 0.0014 - val_mae: 0.0365\nEpoch 108/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0037 - mae: 0.0531 - val_loss: 0.0016 - val_mae: 0.0405\nEpoch 109/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0036 - mae: 0.0517 - val_loss: 0.0016 - val_mae: 0.0392\nEpoch 110/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0037 - mae: 0.0528 - val_loss: 0.0015 - val_mae: 0.0385\nEpoch 111/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0038 - mae: 0.0530 - val_loss: 0.0015 - val_mae: 0.0371\nEpoch 112/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0037 - mae: 0.0520 - val_loss: 0.0014 - val_mae: 0.0365\nEpoch 113/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0036 - mae: 0.0520 - val_loss: 0.0015 - val_mae: 0.0387\nEpoch 114/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0035 - mae: 0.0513 - val_loss: 0.0016 - val_mae: 0.0393\nEpoch 115/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0037 - mae: 0.0520 - val_loss: 0.0015 - val_mae: 0.0381\nEpoch 116/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0037 - mae: 0.0527 - val_loss: 0.0016 - val_mae: 0.0400\nEpoch 117/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0037 - mae: 0.0523 - val_loss: 0.0014 - val_mae: 0.0358\nEpoch 118/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0037 - mae: 0.0529 - val_loss: 0.0015 - val_mae: 0.0381\nEpoch 119/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0038 - mae: 0.0527 - val_loss: 0.0015 - val_mae: 0.0378\nEpoch 120/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0034 - mae: 0.0508 - val_loss: 0.0014 - val_mae: 0.0360\nEpoch 121/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0035 - mae: 0.0507 - val_loss: 0.0014 - val_mae: 0.0367\nEpoch 122/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0036 - mae: 0.0526 - val_loss: 0.0015 - val_mae: 0.0380\nEpoch 123/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0036 - mae: 0.0516 - val_loss: 0.0015 - val_mae: 0.0385\nEpoch 124/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0036 - mae: 0.0518 - val_loss: 0.0014 - val_mae: 0.0362\nEpoch 125/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0036 - mae: 0.0514 - val_loss: 0.0015 - val_mae: 0.0382\nEpoch 126/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0036 - mae: 0.0511 - val_loss: 0.0015 - val_mae: 0.0387\nEpoch 127/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0037 - mae: 0.0516 - val_loss: 0.0015 - val_mae: 0.0372\nEpoch 128/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0036 - mae: 0.0514 - val_loss: 0.0015 - val_mae: 0.0370\nEpoch 129/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0035 - mae: 0.0505 - val_loss: 0.0015 - val_mae: 0.0375\nEpoch 130/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0037 - mae: 0.0516 - val_loss: 0.0014 - val_mae: 0.0360\nEpoch 131/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0036 - mae: 0.0517 - val_loss: 0.0015 - val_mae: 0.0384\nEpoch 132/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0035 - mae: 0.0506 - val_loss: 0.0014 - val_mae: 0.0367\nEpoch 133/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0037 - mae: 0.0516 - val_loss: 0.0016 - val_mae: 0.0399\nEpoch 134/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0036 - mae: 0.0517 - val_loss: 0.0015 - val_mae: 0.0368\nEpoch 135/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0037 - mae: 0.0521 - val_loss: 0.0015 - val_mae: 0.0374\nEpoch 136/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0036 - mae: 0.0517 - val_loss: 0.0014 - val_mae: 0.0359\nEpoch 137/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0035 - mae: 0.0507 - val_loss: 0.0014 - val_mae: 0.0354\nEpoch 138/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0035 - mae: 0.0509 - val_loss: 0.0014 - val_mae: 0.0369\nEpoch 139/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0035 - mae: 0.0507 - val_loss: 0.0015 - val_mae: 0.0372\nEpoch 140/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0036 - mae: 0.0511 - val_loss: 0.0015 - val_mae: 0.0380\nEpoch 141/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0037 - mae: 0.0518 - val_loss: 0.0014 - val_mae: 0.0355\nEpoch 142/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0034 - mae: 0.0503 - val_loss: 0.0013 - val_mae: 0.0345\nEpoch 143/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0036 - mae: 0.0516 - val_loss: 0.0015 - val_mae: 0.0376\nEpoch 144/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0035 - mae: 0.0505 - val_loss: 0.0015 - val_mae: 0.0375\nEpoch 145/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0035 - mae: 0.0506 - val_loss: 0.0013 - val_mae: 0.0344\nEpoch 146/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 15ms/step - loss: 0.0035 - mae: 0.0515 - val_loss: 0.0014 - val_mae: 0.0352\nEpoch 147/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0034 - mae: 0.0505 - val_loss: 0.0015 - val_mae: 0.0372\nEpoch 148/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0037 - mae: 0.0512 - val_loss: 0.0014 - val_mae: 0.0367\nEpoch 149/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0035 - mae: 0.0501 - val_loss: 0.0014 - val_mae: 0.0348\nEpoch 150/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0033 - mae: 0.0496 - val_loss: 0.0014 - val_mae: 0.0360\nEpoch 151/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0035 - mae: 0.0498 - val_loss: 0.0014 - val_mae: 0.0362\nEpoch 152/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0034 - mae: 0.0491 - val_loss: 0.0014 - val_mae: 0.0353\nEpoch 153/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0035 - mae: 0.0504 - val_loss: 0.0014 - val_mae: 0.0361\nEpoch 154/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0035 - mae: 0.0504 - val_loss: 0.0014 - val_mae: 0.0360\nEpoch 155/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0034 - mae: 0.0495 - val_loss: 0.0014 - val_mae: 0.0351\nEpoch 156/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0037 - mae: 0.0517 - val_loss: 0.0015 - val_mae: 0.0369\nEpoch 157/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0034 - mae: 0.0494 - val_loss: 0.0014 - val_mae: 0.0351\nEpoch 158/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0037 - mae: 0.0514 - val_loss: 0.0014 - val_mae: 0.0364\nEpoch 159/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0035 - mae: 0.0498 - val_loss: 0.0016 - val_mae: 0.0390\nEpoch 160/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0034 - mae: 0.0496 - val_loss: 0.0014 - val_mae: 0.0359\nEpoch 161/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0035 - mae: 0.0507 - val_loss: 0.0015 - val_mae: 0.0366\nEpoch 162/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0035 - mae: 0.0497 - val_loss: 0.0014 - val_mae: 0.0351\nEpoch 163/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0035 - mae: 0.0503 - val_loss: 0.0014 - val_mae: 0.0358\nEpoch 164/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0034 - mae: 0.0496 - val_loss: 0.0014 - val_mae: 0.0352\nEpoch 165/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0034 - mae: 0.0497 - val_loss: 0.0014 - val_mae: 0.0356\nEpoch 166/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0035 - mae: 0.0489 - val_loss: 0.0014 - val_mae: 0.0349\nEpoch 167/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0035 - mae: 0.0501 - val_loss: 0.0014 - val_mae: 0.0349\nEpoch 168/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0035 - mae: 0.0503 - val_loss: 0.0014 - val_mae: 0.0352\nEpoch 169/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0034 - mae: 0.0493 - val_loss: 0.0014 - val_mae: 0.0358\nEpoch 170/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0034 - mae: 0.0492 - val_loss: 0.0015 - val_mae: 0.0369\nEpoch 171/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0034 - mae: 0.0495 - val_loss: 0.0014 - val_mae: 0.0363\nEpoch 172/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0033 - mae: 0.0493 - val_loss: 0.0014 - val_mae: 0.0352\nEpoch 173/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0035 - mae: 0.0506 - val_loss: 0.0013 - val_mae: 0.0342\nEpoch 174/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0035 - mae: 0.0505 - val_loss: 0.0014 - val_mae: 0.0364\nEpoch 175/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0036 - mae: 0.0510 - val_loss: 0.0014 - val_mae: 0.0358\nEpoch 176/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0033 - mae: 0.0486 - val_loss: 0.0014 - val_mae: 0.0347\nEpoch 177/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0033 - mae: 0.0486 - val_loss: 0.0014 - val_mae: 0.0348\nEpoch 178/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0033 - mae: 0.0493 - val_loss: 0.0014 - val_mae: 0.0362\nEpoch 179/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0034 - mae: 0.0490 - val_loss: 0.0013 - val_mae: 0.0337\nEpoch 180/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0035 - mae: 0.0503 - val_loss: 0.0014 - val_mae: 0.0349\nEpoch 181/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0034 - mae: 0.0499 - val_loss: 0.0014 - val_mae: 0.0354\nEpoch 182/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0036 - mae: 0.0500 - val_loss: 0.0014 - val_mae: 0.0348\nEpoch 183/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0035 - mae: 0.0501 - val_loss: 0.0014 - val_mae: 0.0350\nEpoch 184/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0034 - mae: 0.0490 - val_loss: 0.0014 - val_mae: 0.0356\nEpoch 185/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0035 - mae: 0.0501 - val_loss: 0.0015 - val_mae: 0.0369\nEpoch 186/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0036 - mae: 0.0502 - val_loss: 0.0014 - val_mae: 0.0357\nEpoch 187/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0037 - mae: 0.0502 - val_loss: 0.0014 - val_mae: 0.0342\nEpoch 188/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0032 - mae: 0.0484 - val_loss: 0.0013 - val_mae: 0.0336\nEpoch 189/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0033 - mae: 0.0490 - val_loss: 0.0014 - val_mae: 0.0350\nEpoch 190/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0032 - mae: 0.0481 - val_loss: 0.0013 - val_mae: 0.0342\nEpoch 191/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0034 - mae: 0.0489 - val_loss: 0.0014 - val_mae: 0.0354\nEpoch 192/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0034 - mae: 0.0491 - val_loss: 0.0014 - val_mae: 0.0352\nEpoch 193/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0033 - mae: 0.0488 - val_loss: 0.0014 - val_mae: 0.0348\nEpoch 194/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0035 - mae: 0.0494 - val_loss: 0.0014 - val_mae: 0.0356\nEpoch 195/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0031 - mae: 0.0475 - val_loss: 0.0013 - val_mae: 0.0340\nEpoch 196/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0034 - mae: 0.0496 - val_loss: 0.0014 - val_mae: 0.0349\nEpoch 197/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0034 - mae: 0.0489 - val_loss: 0.0014 - val_mae: 0.0349\nEpoch 198/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0035 - mae: 0.0499 - val_loss: 0.0014 - val_mae: 0.0348\nEpoch 199/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0034 - mae: 0.0494 - val_loss: 0.0013 - val_mae: 0.0342\nEpoch 200/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0034 - mae: 0.0494 - val_loss: 0.0014 - val_mae: 0.0354\nEpoch 201/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0035 - mae: 0.0494 - val_loss: 0.0015 - val_mae: 0.0371\nEpoch 202/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0033 - mae: 0.0483 - val_loss: 0.0014 - val_mae: 0.0345\nEpoch 203/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0033 - mae: 0.0485 - val_loss: 0.0014 - val_mae: 0.0353\nEpoch 204/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0033 - mae: 0.0490 - val_loss: 0.0014 - val_mae: 0.0354\nEpoch 205/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0033 - mae: 0.0482 - val_loss: 0.0014 - val_mae: 0.0348\nEpoch 206/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0032 - mae: 0.0483 - val_loss: 0.0014 - val_mae: 0.0351\nEpoch 207/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0034 - mae: 0.0490 - val_loss: 0.0014 - val_mae: 0.0357\nEpoch 208/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 18ms/step - loss: 0.0032 - mae: 0.0474 - val_loss: 0.0014 - val_mae: 0.0347\nEpoch 209/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0031 - mae: 0.0477 - val_loss: 0.0013 - val_mae: 0.0339\nEpoch 210/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0034 - mae: 0.0495 - val_loss: 0.0013 - val_mae: 0.0342\nEpoch 211/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0032 - mae: 0.0476 - val_loss: 0.0013 - val_mae: 0.0338\nEpoch 212/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0034 - mae: 0.0499 - val_loss: 0.0014 - val_mae: 0.0357\nEpoch 213/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0033 - mae: 0.0486 - val_loss: 0.0014 - val_mae: 0.0353\nEpoch 214/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0034 - mae: 0.0487 - val_loss: 0.0014 - val_mae: 0.0351\nEpoch 215/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0032 - mae: 0.0480 - val_loss: 0.0014 - val_mae: 0.0349\nEpoch 216/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0033 - mae: 0.0487 - val_loss: 0.0013 - val_mae: 0.0344\nEpoch 217/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0034 - mae: 0.0493 - val_loss: 0.0014 - val_mae: 0.0350\nEpoch 218/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0032 - mae: 0.0477 - val_loss: 0.0013 - val_mae: 0.0343\nEpoch 219/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0032 - mae: 0.0483 - val_loss: 0.0013 - val_mae: 0.0339\nEpoch 220/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0031 - mae: 0.0466 - val_loss: 0.0013 - val_mae: 0.0339\nEpoch 221/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0031 - mae: 0.0470 - val_loss: 0.0014 - val_mae: 0.0361\nEpoch 222/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0033 - mae: 0.0484 - val_loss: 0.0014 - val_mae: 0.0355\nEpoch 223/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0033 - mae: 0.0484 - val_loss: 0.0013 - val_mae: 0.0345\nEpoch 224/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0030 - mae: 0.0460 - val_loss: 0.0013 - val_mae: 0.0339\nEpoch 225/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0032 - mae: 0.0481 - val_loss: 0.0014 - val_mae: 0.0345\nEpoch 226/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0033 - mae: 0.0482 - val_loss: 0.0014 - val_mae: 0.0348\nEpoch 227/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0033 - mae: 0.0480 - val_loss: 0.0014 - val_mae: 0.0346\nEpoch 228/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0034 - mae: 0.0481 - val_loss: 0.0014 - val_mae: 0.0349\nEpoch 229/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0033 - mae: 0.0484 - val_loss: 0.0013 - val_mae: 0.0341\nEpoch 230/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0031 - mae: 0.0477 - val_loss: 0.0014 - val_mae: 0.0348\nEpoch 231/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0034 - mae: 0.0495 - val_loss: 0.0013 - val_mae: 0.0344\nEpoch 232/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0033 - mae: 0.0482 - val_loss: 0.0013 - val_mae: 0.0341\nEpoch 233/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 16ms/step - loss: 0.0033 - mae: 0.0476 - val_loss: 0.0014 - val_mae: 0.0348\nEpoch 234/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0032 - mae: 0.0480 - val_loss: 0.0013 - val_mae: 0.0342\nEpoch 235/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0031 - mae: 0.0470 - val_loss: 0.0014 - val_mae: 0.0357\nEpoch 236/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0034 - mae: 0.0489 - val_loss: 0.0013 - val_mae: 0.0339\nEpoch 237/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0032 - mae: 0.0483 - val_loss: 0.0013 - val_mae: 0.0343\nEpoch 238/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 17ms/step - loss: 0.0033 - mae: 0.0486 - val_loss: 0.0013 - val_mae: 0.0341\n\u001b[1m3/3\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 8ms/step - loss: 0.0012 - mae: 0.0337 \n","output_type":"stream"},{"execution_count":60,"output_type":"execute_result","data":{"text/plain":"33648.80383014679"},"metadata":{}}]},{"cell_type":"markdown","source":"## RNN to predict at each time step.","metadata":{}},{"cell_type":"code","source":"my_series = tf.data.Dataset.range(7)\ndataset = to_windows(to_windows(my_series, 3), 4)\nlist(dataset)","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:42:45.456671Z","iopub.execute_input":"2024-07-11T18:42:45.457375Z","iopub.status.idle":"2024-07-11T18:42:45.572147Z","shell.execute_reply.started":"2024-07-11T18:42:45.457327Z","shell.execute_reply":"2024-07-11T18:42:45.569526Z"},"trusted":true},"execution_count":61,"outputs":[{"execution_count":61,"output_type":"execute_result","data":{"text/plain":"[<tf.Tensor: shape=(4, 3), dtype=int64, numpy=\n array([[0, 1, 2],\n        [1, 2, 3],\n        [2, 3, 4],\n        [3, 4, 5]])>,\n <tf.Tensor: shape=(4, 3), dtype=int64, numpy=\n array([[1, 2, 3],\n        [2, 3, 4],\n        [3, 4, 5],\n        [4, 5, 6]])>]"},"metadata":{}}]},{"cell_type":"code","source":"dataset = dataset.map(lambda S: (S[:, 0], S[:, 1:]))\nlist(dataset)","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:42:55.978687Z","iopub.execute_input":"2024-07-11T18:42:55.979239Z","iopub.status.idle":"2024-07-11T18:42:56.084423Z","shell.execute_reply.started":"2024-07-11T18:42:55.979166Z","shell.execute_reply":"2024-07-11T18:42:56.082548Z"},"trusted":true},"execution_count":62,"outputs":[{"execution_count":62,"output_type":"execute_result","data":{"text/plain":"[(<tf.Tensor: shape=(4,), dtype=int64, numpy=array([0, 1, 2, 3])>,\n  <tf.Tensor: shape=(4, 2), dtype=int64, numpy=\n  array([[1, 2],\n         [2, 3],\n         [3, 4],\n         [4, 5]])>),\n (<tf.Tensor: shape=(4,), dtype=int64, numpy=array([1, 2, 3, 4])>,\n  <tf.Tensor: shape=(4, 2), dtype=int64, numpy=\n  array([[2, 3],\n         [3, 4],\n         [4, 5],\n         [5, 6]])>)]"},"metadata":{}}]},{"cell_type":"code","source":"def to_seq2seq_dataset(series, seq_length=56, ahead=14, target_col=1,\n                       batch_size=32, shuffle=False, seed=None):\n    ds = to_windows(tf.data.Dataset.from_tensor_slices(series), ahead + 1)\n    ds = to_windows(ds, seq_length).map(lambda S: (S[:, 0], S[:, 1:, 1]))\n    if shuffle:\n        ds = ds.shuffle(8 * batch_size, seed=seed)\n    return ds.batch(batch_size)","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:43:07.326476Z","iopub.execute_input":"2024-07-11T18:43:07.326984Z","iopub.status.idle":"2024-07-11T18:43:07.336673Z","shell.execute_reply.started":"2024-07-11T18:43:07.326947Z","shell.execute_reply":"2024-07-11T18:43:07.335202Z"},"trusted":true},"execution_count":63,"outputs":[]},{"cell_type":"code","source":"seq2seq_train = to_seq2seq_dataset(mulvar_train, shuffle=True, seed=42)\nseq2seq_valid = to_seq2seq_dataset(mulvar_valid)","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:43:16.986389Z","iopub.execute_input":"2024-07-11T18:43:16.987655Z","iopub.status.idle":"2024-07-11T18:43:17.166849Z","shell.execute_reply.started":"2024-07-11T18:43:16.987595Z","shell.execute_reply":"2024-07-11T18:43:17.165312Z"},"trusted":true},"execution_count":64,"outputs":[]},{"cell_type":"code","source":"tf.random.set_seed(42)  # ensures reproducibility\nseq2seq_model = tf.keras.Sequential([\n    tf.keras.layers.SimpleRNN(32, return_sequences=True, input_shape=[None, 5]),\n    tf.keras.layers.Dense(14)\n    # equivalent: tf.keras.layers.TimeDistributed(tf.keras.layers.Dense(14))\n    # also equivalent: tf.keras.layers.Conv1D(14, kernel_size=1)\n])","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:43:32.969170Z","iopub.execute_input":"2024-07-11T18:43:32.969667Z","iopub.status.idle":"2024-07-11T18:43:33.060752Z","shell.execute_reply.started":"2024-07-11T18:43:32.969634Z","shell.execute_reply":"2024-07-11T18:43:33.059194Z"},"trusted":true},"execution_count":65,"outputs":[{"name":"stderr","text":"/opt/conda/lib/python3.10/site-packages/keras/src/layers/rnn/rnn.py:204: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n  super().__init__(**kwargs)\n","output_type":"stream"}]},{"cell_type":"code","source":"fit_and_evaluate(seq2seq_model, seq2seq_train, seq2seq_valid,\n                 learning_rate=0.1)","metadata":{"execution":{"iopub.status.busy":"2024-07-11T18:43:42.466897Z","iopub.execute_input":"2024-07-11T18:43:42.467564Z","iopub.status.idle":"2024-07-11T18:48:40.448504Z","shell.execute_reply.started":"2024-07-11T18:43:42.467515Z","shell.execute_reply":"2024-07-11T18:48:40.445620Z"},"trusted":true},"execution_count":66,"outputs":[{"name":"stdout","text":"Epoch 1/500\n     31/Unknown \u001b[1m3s\u001b[0m 24ms/step - loss: 0.1262 - mae: 0.3810","output_type":"stream"},{"name":"stderr","text":"/opt/conda/lib/python3.10/contextlib.py:153: UserWarning: Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches. You may need to use the `.repeat()` function when building your dataset.\n  self.gen.throw(typ, value, traceback)\n","output_type":"stream"},{"name":"stdout","text":"\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 38ms/step - loss: 0.1203 - mae: 0.3695 - val_loss: 0.0153 - val_mae: 0.1335\nEpoch 2/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0119 - mae: 0.1196 - val_loss: 0.0080 - val_mae: 0.0932\nEpoch 3/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0080 - mae: 0.0933 - val_loss: 0.0058 - val_mae: 0.0757\nEpoch 4/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0066 - mae: 0.0825 - val_loss: 0.0054 - val_mae: 0.0687\nEpoch 5/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0055 - mae: 0.0713 - val_loss: 0.0050 - val_mae: 0.0653\nEpoch 6/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0049 - mae: 0.0663 - val_loss: 0.0045 - val_mae: 0.0594\nEpoch 7/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0048 - mae: 0.0647 - val_loss: 0.0047 - val_mae: 0.0628\nEpoch 8/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0046 - mae: 0.0621 - val_loss: 0.0045 - val_mae: 0.0596\nEpoch 9/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0044 - mae: 0.0607 - val_loss: 0.0041 - val_mae: 0.0548\nEpoch 10/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0043 - mae: 0.0603 - val_loss: 0.0039 - val_mae: 0.0532\nEpoch 11/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0043 - mae: 0.0607 - val_loss: 0.0043 - val_mae: 0.0575\nEpoch 12/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0042 - mae: 0.0592 - val_loss: 0.0042 - val_mae: 0.0566\nEpoch 13/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0041 - mae: 0.0582 - val_loss: 0.0040 - val_mae: 0.0546\nEpoch 14/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0041 - mae: 0.0580 - val_loss: 0.0042 - val_mae: 0.0572\nEpoch 15/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0040 - mae: 0.0566 - val_loss: 0.0043 - val_mae: 0.0583\nEpoch 16/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0039 - mae: 0.0563 - val_loss: 0.0039 - val_mae: 0.0533\nEpoch 17/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0039 - mae: 0.0572 - val_loss: 0.0040 - val_mae: 0.0545\nEpoch 18/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0039 - mae: 0.0564 - val_loss: 0.0039 - val_mae: 0.0537\nEpoch 19/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0039 - mae: 0.0557 - val_loss: 0.0040 - val_mae: 0.0542\nEpoch 20/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0038 - mae: 0.0555 - val_loss: 0.0040 - val_mae: 0.0541\nEpoch 21/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0038 - mae: 0.0549 - val_loss: 0.0039 - val_mae: 0.0530\nEpoch 22/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0038 - mae: 0.0554 - val_loss: 0.0040 - val_mae: 0.0545\nEpoch 23/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0038 - mae: 0.0550 - val_loss: 0.0038 - val_mae: 0.0517\nEpoch 24/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0038 - mae: 0.0558 - val_loss: 0.0039 - val_mae: 0.0529\nEpoch 25/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0037 - mae: 0.0553 - val_loss: 0.0037 - val_mae: 0.0507\nEpoch 26/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0037 - mae: 0.0553 - val_loss: 0.0039 - val_mae: 0.0526\nEpoch 27/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 28ms/step - loss: 0.0036 - mae: 0.0539 - val_loss: 0.0039 - val_mae: 0.0528\nEpoch 28/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0036 - mae: 0.0536 - val_loss: 0.0038 - val_mae: 0.0516\nEpoch 29/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - loss: 0.0036 - mae: 0.0538 - val_loss: 0.0037 - val_mae: 0.0501\nEpoch 30/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 26ms/step - loss: 0.0037 - mae: 0.0548 - val_loss: 0.0037 - val_mae: 0.0501\nEpoch 31/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0036 - mae: 0.0533 - val_loss: 0.0036 - val_mae: 0.0490\nEpoch 32/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0036 - mae: 0.0541 - val_loss: 0.0038 - val_mae: 0.0513\nEpoch 33/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0036 - mae: 0.0531 - val_loss: 0.0037 - val_mae: 0.0503\nEpoch 34/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0035 - mae: 0.0529 - val_loss: 0.0037 - val_mae: 0.0498\nEpoch 35/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - loss: 0.0035 - mae: 0.0529 - val_loss: 0.0037 - val_mae: 0.0504\nEpoch 36/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0036 - mae: 0.0537 - val_loss: 0.0037 - val_mae: 0.0510\nEpoch 37/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0035 - mae: 0.0523 - val_loss: 0.0036 - val_mae: 0.0485\nEpoch 38/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0037 - mae: 0.0549 - val_loss: 0.0037 - val_mae: 0.0507\nEpoch 39/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0035 - mae: 0.0529 - val_loss: 0.0037 - val_mae: 0.0499\nEpoch 40/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0035 - mae: 0.0525 - val_loss: 0.0037 - val_mae: 0.0500\nEpoch 41/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0035 - mae: 0.0529 - val_loss: 0.0037 - val_mae: 0.0498\nEpoch 42/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0035 - mae: 0.0529 - val_loss: 0.0036 - val_mae: 0.0488\nEpoch 43/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0035 - mae: 0.0533 - val_loss: 0.0036 - val_mae: 0.0484\nEpoch 44/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0035 - mae: 0.0530 - val_loss: 0.0037 - val_mae: 0.0510\nEpoch 45/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0034 - mae: 0.0516 - val_loss: 0.0036 - val_mae: 0.0495\nEpoch 46/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0034 - mae: 0.0518 - val_loss: 0.0036 - val_mae: 0.0489\nEpoch 47/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0036 - mae: 0.0536 - val_loss: 0.0036 - val_mae: 0.0492\nEpoch 48/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0035 - mae: 0.0528 - val_loss: 0.0036 - val_mae: 0.0492\nEpoch 49/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0035 - mae: 0.0523 - val_loss: 0.0036 - val_mae: 0.0486\nEpoch 50/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0035 - mae: 0.0528 - val_loss: 0.0036 - val_mae: 0.0486\nEpoch 51/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0035 - mae: 0.0527 - val_loss: 0.0036 - val_mae: 0.0493\nEpoch 52/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0035 - mae: 0.0523 - val_loss: 0.0037 - val_mae: 0.0499\nEpoch 53/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0033 - mae: 0.0514 - val_loss: 0.0036 - val_mae: 0.0486\nEpoch 54/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0034 - mae: 0.0518 - val_loss: 0.0037 - val_mae: 0.0500\nEpoch 55/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0033 - mae: 0.0511 - val_loss: 0.0037 - val_mae: 0.0496\nEpoch 56/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0034 - mae: 0.0522 - val_loss: 0.0038 - val_mae: 0.0515\nEpoch 57/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0033 - mae: 0.0507 - val_loss: 0.0036 - val_mae: 0.0482\nEpoch 58/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 29ms/step - loss: 0.0034 - mae: 0.0520 - val_loss: 0.0036 - val_mae: 0.0488\nEpoch 59/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0034 - mae: 0.0518 - val_loss: 0.0036 - val_mae: 0.0483\nEpoch 60/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0034 - mae: 0.0519 - val_loss: 0.0037 - val_mae: 0.0505\nEpoch 61/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - loss: 0.0033 - mae: 0.0502 - val_loss: 0.0035 - val_mae: 0.0478\nEpoch 62/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 26ms/step - loss: 0.0035 - mae: 0.0534 - val_loss: 0.0036 - val_mae: 0.0481\nEpoch 63/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 28ms/step - loss: 0.0034 - mae: 0.0522 - val_loss: 0.0036 - val_mae: 0.0488\nEpoch 64/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 26ms/step - loss: 0.0034 - mae: 0.0522 - val_loss: 0.0036 - val_mae: 0.0484\nEpoch 65/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0034 - mae: 0.0518 - val_loss: 0.0037 - val_mae: 0.0495\nEpoch 66/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0033 - mae: 0.0509 - val_loss: 0.0037 - val_mae: 0.0505\nEpoch 67/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0034 - mae: 0.0515 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 68/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 31ms/step - loss: 0.0034 - mae: 0.0525 - val_loss: 0.0037 - val_mae: 0.0493\nEpoch 69/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 24ms/step - loss: 0.0034 - mae: 0.0511 - val_loss: 0.0036 - val_mae: 0.0483\nEpoch 70/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0033 - mae: 0.0509 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 71/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0034 - mae: 0.0514 - val_loss: 0.0036 - val_mae: 0.0484\nEpoch 72/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - loss: 0.0034 - mae: 0.0513 - val_loss: 0.0036 - val_mae: 0.0482\nEpoch 73/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0034 - mae: 0.0520 - val_loss: 0.0036 - val_mae: 0.0482\nEpoch 74/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0033 - mae: 0.0511 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 75/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0033 - mae: 0.0523 - val_loss: 0.0036 - val_mae: 0.0482\nEpoch 76/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0034 - mae: 0.0522 - val_loss: 0.0036 - val_mae: 0.0482\nEpoch 77/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0034 - mae: 0.0522 - val_loss: 0.0036 - val_mae: 0.0483\nEpoch 78/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0034 - mae: 0.0534 - val_loss: 0.0036 - val_mae: 0.0483\nEpoch 79/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0033 - mae: 0.0511 - val_loss: 0.0036 - val_mae: 0.0481\nEpoch 80/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0033 - mae: 0.0502 - val_loss: 0.0036 - val_mae: 0.0486\nEpoch 81/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0033 - mae: 0.0509 - val_loss: 0.0036 - val_mae: 0.0480\nEpoch 82/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0033 - mae: 0.0513 - val_loss: 0.0036 - val_mae: 0.0486\nEpoch 83/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0034 - mae: 0.0516 - val_loss: 0.0036 - val_mae: 0.0486\nEpoch 84/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0033 - mae: 0.0503 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 85/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 30ms/step - loss: 0.0032 - mae: 0.0504 - val_loss: 0.0036 - val_mae: 0.0480\nEpoch 86/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 35ms/step - loss: 0.0033 - mae: 0.0511 - val_loss: 0.0037 - val_mae: 0.0493\nEpoch 87/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 32ms/step - loss: 0.0032 - mae: 0.0497 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 88/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 30ms/step - loss: 0.0033 - mae: 0.0507 - val_loss: 0.0036 - val_mae: 0.0480\nEpoch 89/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 31ms/step - loss: 0.0033 - mae: 0.0503 - val_loss: 0.0036 - val_mae: 0.0484\nEpoch 90/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 26ms/step - loss: 0.0032 - mae: 0.0502 - val_loss: 0.0036 - val_mae: 0.0482\nEpoch 91/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 26ms/step - loss: 0.0032 - mae: 0.0499 - val_loss: 0.0036 - val_mae: 0.0484\nEpoch 92/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0033 - mae: 0.0504 - val_loss: 0.0036 - val_mae: 0.0485\nEpoch 93/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0033 - mae: 0.0504 - val_loss: 0.0036 - val_mae: 0.0481\nEpoch 94/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0033 - mae: 0.0511 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 95/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0035 - mae: 0.0536 - val_loss: 0.0036 - val_mae: 0.0480\nEpoch 96/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0033 - mae: 0.0506 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 97/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 26ms/step - loss: 0.0033 - mae: 0.0512 - val_loss: 0.0036 - val_mae: 0.0482\nEpoch 98/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0034 - mae: 0.0527 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 99/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0033 - mae: 0.0504 - val_loss: 0.0036 - val_mae: 0.0481\nEpoch 100/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0033 - mae: 0.0501 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 101/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0033 - mae: 0.0509 - val_loss: 0.0036 - val_mae: 0.0480\nEpoch 102/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0032 - mae: 0.0505 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 103/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0033 - mae: 0.0505 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 104/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0033 - mae: 0.0505 - val_loss: 0.0036 - val_mae: 0.0482\nEpoch 105/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0033 - mae: 0.0502 - val_loss: 0.0036 - val_mae: 0.0482\nEpoch 106/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0032 - mae: 0.0505 - val_loss: 0.0036 - val_mae: 0.0480\nEpoch 107/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0032 - mae: 0.0497 - val_loss: 0.0036 - val_mae: 0.0482\nEpoch 108/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0032 - mae: 0.0493 - val_loss: 0.0037 - val_mae: 0.0486\nEpoch 109/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0032 - mae: 0.0492 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 110/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0032 - mae: 0.0499 - val_loss: 0.0036 - val_mae: 0.0481\nEpoch 111/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0033 - mae: 0.0503 - val_loss: 0.0037 - val_mae: 0.0488\nEpoch 112/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0033 - mae: 0.0505 - val_loss: 0.0036 - val_mae: 0.0480\nEpoch 113/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0032 - mae: 0.0504 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 114/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0032 - mae: 0.0504 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 115/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0032 - mae: 0.0497 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 116/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 28ms/step - loss: 0.0033 - mae: 0.0507 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 117/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 31ms/step - loss: 0.0032 - mae: 0.0495 - val_loss: 0.0037 - val_mae: 0.0490\nEpoch 118/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 29ms/step - loss: 0.0031 - mae: 0.0485 - val_loss: 0.0036 - val_mae: 0.0478\nEpoch 119/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0032 - mae: 0.0501 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 120/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - loss: 0.0033 - mae: 0.0507 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 121/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0033 - mae: 0.0510 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 122/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0032 - mae: 0.0498 - val_loss: 0.0036 - val_mae: 0.0483\nEpoch 123/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 26ms/step - loss: 0.0032 - mae: 0.0492 - val_loss: 0.0036 - val_mae: 0.0482\nEpoch 124/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0486 - val_loss: 0.0037 - val_mae: 0.0484\nEpoch 125/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0487 - val_loss: 0.0036 - val_mae: 0.0480\nEpoch 126/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0031 - mae: 0.0486 - val_loss: 0.0038 - val_mae: 0.0499\nEpoch 127/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0032 - mae: 0.0486 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 128/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0032 - mae: 0.0501 - val_loss: 0.0037 - val_mae: 0.0482\nEpoch 129/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0487 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 130/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0032 - mae: 0.0491 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 131/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 26ms/step - loss: 0.0031 - mae: 0.0484 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 132/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0031 - mae: 0.0494 - val_loss: 0.0037 - val_mae: 0.0482\nEpoch 133/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0032 - mae: 0.0498 - val_loss: 0.0037 - val_mae: 0.0486\nEpoch 134/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0032 - mae: 0.0497 - val_loss: 0.0037 - val_mae: 0.0484\nEpoch 135/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0032 - mae: 0.0492 - val_loss: 0.0036 - val_mae: 0.0481\nEpoch 136/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0032 - mae: 0.0490 - val_loss: 0.0036 - val_mae: 0.0478\nEpoch 137/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0032 - mae: 0.0493 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 138/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0033 - mae: 0.0500 - val_loss: 0.0037 - val_mae: 0.0483\nEpoch 139/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0484 - val_loss: 0.0037 - val_mae: 0.0486\nEpoch 140/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0487 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 141/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0493 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 142/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0490 - val_loss: 0.0037 - val_mae: 0.0484\nEpoch 143/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0032 - mae: 0.0497 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 144/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0032 - mae: 0.0503 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 145/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0032 - mae: 0.0504 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 146/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0482 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 147/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0487 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 148/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0030 - mae: 0.0479 - val_loss: 0.0037 - val_mae: 0.0480\nEpoch 149/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0032 - mae: 0.0491 - val_loss: 0.0036 - val_mae: 0.0478\nEpoch 150/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0032 - mae: 0.0501 - val_loss: 0.0036 - val_mae: 0.0480\nEpoch 151/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - loss: 0.0031 - mae: 0.0485 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 152/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0031 - mae: 0.0494 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 153/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0031 - mae: 0.0488 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 154/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0032 - mae: 0.0507 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 155/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0032 - mae: 0.0496 - val_loss: 0.0037 - val_mae: 0.0486\nEpoch 156/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0483 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 157/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0480 - val_loss: 0.0037 - val_mae: 0.0485\nEpoch 158/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0477 - val_loss: 0.0037 - val_mae: 0.0486\nEpoch 159/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0478 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 160/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0480 - val_loss: 0.0036 - val_mae: 0.0478\nEpoch 161/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0031 - mae: 0.0479 - val_loss: 0.0037 - val_mae: 0.0482\nEpoch 162/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0478 - val_loss: 0.0037 - val_mae: 0.0480\nEpoch 163/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0491 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 164/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0032 - mae: 0.0492 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 165/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0485 - val_loss: 0.0037 - val_mae: 0.0484\nEpoch 166/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0031 - mae: 0.0480 - val_loss: 0.0037 - val_mae: 0.0486\nEpoch 167/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0473 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 168/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0479 - val_loss: 0.0036 - val_mae: 0.0478\nEpoch 169/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0476 - val_loss: 0.0037 - val_mae: 0.0478\nEpoch 170/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0480 - val_loss: 0.0037 - val_mae: 0.0482\nEpoch 171/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0481 - val_loss: 0.0037 - val_mae: 0.0482\nEpoch 172/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0485 - val_loss: 0.0037 - val_mae: 0.0488\nEpoch 173/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0491 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 174/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0492 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 175/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0480 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 176/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0483 - val_loss: 0.0037 - val_mae: 0.0497\nEpoch 177/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0473 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 178/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0482 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 179/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0493 - val_loss: 0.0036 - val_mae: 0.0471\nEpoch 180/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0032 - mae: 0.0500 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 181/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - loss: 0.0032 - mae: 0.0491 - val_loss: 0.0036 - val_mae: 0.0472\nEpoch 182/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0030 - mae: 0.0477 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 183/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0478 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 184/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0478 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 185/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0481 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 186/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0478 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 187/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0478 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 188/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0029 - mae: 0.0469 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 189/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0482 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 190/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0480 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 191/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0473 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 192/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0480 - val_loss: 0.0036 - val_mae: 0.0480\nEpoch 193/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0474 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 194/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0482 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 195/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0474 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 196/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0480 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 197/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0476 - val_loss: 0.0036 - val_mae: 0.0478\nEpoch 198/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 19ms/step - loss: 0.0031 - mae: 0.0479 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 199/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0030 - mae: 0.0470 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 200/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0475 - val_loss: 0.0037 - val_mae: 0.0484\nEpoch 201/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0471 - val_loss: 0.0036 - val_mae: 0.0480\nEpoch 202/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0029 - mae: 0.0466 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 203/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0473 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 204/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0477 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 205/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0474 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 206/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0475 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 207/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0476 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 208/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0479 - val_loss: 0.0036 - val_mae: 0.0480\nEpoch 209/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0468 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 210/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0473 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 211/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0031 - mae: 0.0484 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 212/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0476 - val_loss: 0.0036 - val_mae: 0.0471\nEpoch 213/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0475 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 214/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0475 - val_loss: 0.0036 - val_mae: 0.0472\nEpoch 215/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0030 - mae: 0.0470 - val_loss: 0.0036 - val_mae: 0.0470\nEpoch 216/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0030 - mae: 0.0475 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 217/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0483 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 218/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0485 - val_loss: 0.0036 - val_mae: 0.0471\nEpoch 219/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0488 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 220/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0031 - mae: 0.0478 - val_loss: 0.0036 - val_mae: 0.0471\nEpoch 221/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0031 - mae: 0.0484 - val_loss: 0.0036 - val_mae: 0.0471\nEpoch 222/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0031 - mae: 0.0480 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 223/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0030 - mae: 0.0481 - val_loss: 0.0036 - val_mae: 0.0470\nEpoch 224/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0029 - mae: 0.0467 - val_loss: 0.0036 - val_mae: 0.0472\nEpoch 225/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0481 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 226/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0030 - mae: 0.0471 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 227/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0475 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 228/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0030 - mae: 0.0472 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 229/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0030 - mae: 0.0473 - val_loss: 0.0037 - val_mae: 0.0484\nEpoch 230/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0477 - val_loss: 0.0036 - val_mae: 0.0487\nEpoch 231/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0491 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 232/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0476 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 233/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0470 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 234/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0472 - val_loss: 0.0036 - val_mae: 0.0480\nEpoch 235/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0030 - mae: 0.0479 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 236/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0029 - mae: 0.0464 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 237/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0030 - mae: 0.0479 - val_loss: 0.0036 - val_mae: 0.0472\nEpoch 238/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0470 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 239/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0031 - mae: 0.0480 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 240/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0475 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 241/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0472 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 242/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0029 - mae: 0.0466 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 243/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0467 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 244/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0030 - mae: 0.0474 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 245/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0464 - val_loss: 0.0036 - val_mae: 0.0478\nEpoch 246/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0030 - mae: 0.0469 - val_loss: 0.0036 - val_mae: 0.0472\nEpoch 247/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0476 - val_loss: 0.0036 - val_mae: 0.0471\nEpoch 248/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - loss: 0.0029 - mae: 0.0465 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 249/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0030 - mae: 0.0464 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 250/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0030 - mae: 0.0471 - val_loss: 0.0036 - val_mae: 0.0478\nEpoch 251/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0030 - mae: 0.0478 - val_loss: 0.0036 - val_mae: 0.0483\nEpoch 252/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0030 - mae: 0.0488 - val_loss: 0.0036 - val_mae: 0.0471\nEpoch 253/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0030 - mae: 0.0481 - val_loss: 0.0035 - val_mae: 0.0471\nEpoch 254/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0030 - mae: 0.0473 - val_loss: 0.0036 - val_mae: 0.0470\nEpoch 255/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0029 - mae: 0.0467 - val_loss: 0.0036 - val_mae: 0.0484\nEpoch 256/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0483 - val_loss: 0.0036 - val_mae: 0.0470\nEpoch 257/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0029 - mae: 0.0460 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 258/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0030 - mae: 0.0476 - val_loss: 0.0036 - val_mae: 0.0471\nEpoch 259/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0029 - mae: 0.0463 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 260/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0029 - mae: 0.0460 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 261/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0030 - mae: 0.0471 - val_loss: 0.0036 - val_mae: 0.0472\nEpoch 262/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0475 - val_loss: 0.0036 - val_mae: 0.0471\nEpoch 263/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0030 - mae: 0.0472 - val_loss: 0.0036 - val_mae: 0.0478\nEpoch 264/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0030 - mae: 0.0472 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 265/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0466 - val_loss: 0.0036 - val_mae: 0.0469\nEpoch 266/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0473 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 267/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0470 - val_loss: 0.0036 - val_mae: 0.0471\nEpoch 268/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0465 - val_loss: 0.0036 - val_mae: 0.0494\nEpoch 269/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0030 - mae: 0.0476 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 270/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0028 - mae: 0.0451 - val_loss: 0.0036 - val_mae: 0.0478\nEpoch 271/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0030 - mae: 0.0469 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 272/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 27ms/step - loss: 0.0030 - mae: 0.0476 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 273/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0029 - mae: 0.0462 - val_loss: 0.0036 - val_mae: 0.0472\nEpoch 274/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0029 - mae: 0.0467 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 275/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0477 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 276/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0475 - val_loss: 0.0035 - val_mae: 0.0471\nEpoch 277/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0467 - val_loss: 0.0036 - val_mae: 0.0472\nEpoch 278/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0471 - val_loss: 0.0036 - val_mae: 0.0478\nEpoch 279/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0029 - mae: 0.0467 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 280/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0467 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 281/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0029 - mae: 0.0462 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 282/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0030 - mae: 0.0468 - val_loss: 0.0036 - val_mae: 0.0472\nEpoch 283/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0029 - mae: 0.0467 - val_loss: 0.0035 - val_mae: 0.0472\nEpoch 284/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0029 - mae: 0.0466 - val_loss: 0.0036 - val_mae: 0.0472\nEpoch 285/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 25ms/step - loss: 0.0029 - mae: 0.0466 - val_loss: 0.0035 - val_mae: 0.0470\nEpoch 286/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0029 - mae: 0.0465 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 287/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0029 - mae: 0.0465 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 288/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0029 - mae: 0.0466 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 289/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0028 - mae: 0.0459 - val_loss: 0.0036 - val_mae: 0.0479\nEpoch 290/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0030 - mae: 0.0472 - val_loss: 0.0036 - val_mae: 0.0488\nEpoch 291/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0029 - mae: 0.0468 - val_loss: 0.0036 - val_mae: 0.0475\nEpoch 292/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0029 - mae: 0.0462 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 293/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0029 - mae: 0.0465 - val_loss: 0.0035 - val_mae: 0.0474\nEpoch 294/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0029 - mae: 0.0463 - val_loss: 0.0037 - val_mae: 0.0501\nEpoch 295/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0029 - mae: 0.0474 - val_loss: 0.0036 - val_mae: 0.0480\nEpoch 296/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0029 - mae: 0.0461 - val_loss: 0.0036 - val_mae: 0.0477\nEpoch 297/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0029 - mae: 0.0456 - val_loss: 0.0036 - val_mae: 0.0478\nEpoch 298/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0029 - mae: 0.0465 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 299/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0029 - mae: 0.0461 - val_loss: 0.0035 - val_mae: 0.0475\nEpoch 300/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0029 - mae: 0.0460 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 301/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0029 - mae: 0.0465 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 302/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0029 - mae: 0.0467 - val_loss: 0.0036 - val_mae: 0.0473\nEpoch 303/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 22ms/step - loss: 0.0029 - mae: 0.0464 - val_loss: 0.0035 - val_mae: 0.0470\nEpoch 304/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0030 - mae: 0.0467 - val_loss: 0.0035 - val_mae: 0.0471\nEpoch 305/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0029 - mae: 0.0460 - val_loss: 0.0036 - val_mae: 0.0489\nEpoch 306/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0029 - mae: 0.0465 - val_loss: 0.0036 - val_mae: 0.0474\nEpoch 307/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 24ms/step - loss: 0.0028 - mae: 0.0456 - val_loss: 0.0035 - val_mae: 0.0472\nEpoch 308/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 26ms/step - loss: 0.0029 - mae: 0.0465 - val_loss: 0.0035 - val_mae: 0.0471\nEpoch 309/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 23ms/step - loss: 0.0029 - mae: 0.0464 - val_loss: 0.0035 - val_mae: 0.0473\nEpoch 310/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0028 - mae: 0.0454 - val_loss: 0.0035 - val_mae: 0.0471\nEpoch 311/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0029 - mae: 0.0461 - val_loss: 0.0035 - val_mae: 0.0476\nEpoch 312/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 20ms/step - loss: 0.0029 - mae: 0.0468 - val_loss: 0.0036 - val_mae: 0.0491\nEpoch 313/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 21ms/step - loss: 0.0029 - mae: 0.0469 - val_loss: 0.0036 - val_mae: 0.0483\nEpoch 314/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 30ms/step - loss: 0.0029 - mae: 0.0460 - val_loss: 0.0036 - val_mae: 0.0476\nEpoch 315/500\n\u001b[1m33/33\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 31ms/step - loss: 0.0029 - mae: 0.0463 - val_loss: 0.0036 - val_mae: 0.0477\n\u001b[1m3/3\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step - loss: 0.0045 - mae: 0.0514\n","output_type":"stream"},{"execution_count":66,"output_type":"execute_result","data":{"text/plain":"46922.25530743599"},"metadata":{}}]}]}