Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Implement model cu 86c1f7e46 #256

Closed
wants to merge 7 commits into from
Closed

Implement model cu 86c1f7e46 #256

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
c077ddc
build: add dev deps
noamsgl Dec 31, 2024
f967851
wip: feat: add forecasts
noamsgl Dec 31, 2024
68751f9
wip: feat: add forecasts
noamsgl Dec 31, 2024
0ddcc48
build: add typer-config
noamsgl Dec 31, 2024
6204a48
feat: finetune within patient and convert to wav
noamsgl Dec 31, 2024
21011e2
feat: add for loop
noamsgl Jan 2, 2025
cd7975a
build: add fastparquet, ruff
noamsgl Jan 2, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
247 changes: 247 additions & 0 deletions notebooks/plot_quantile_forecast.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,247 @@
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import numpy as np\n",
"import webdataset as wds\n",
"import pandas as pd"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"patient_names = [\"pat_96002\"]\n",
"file_paths = [f\"/innovation_cache/merged_eeg/{patient_name}.univariate.tar\" for patient_name in patient_names]\n",
"ds = wds.WebDataset(file_paths, shardshuffle=False).decode(\"l\")\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
"sample = next(ds.iterator())"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"{'__key__': '0',\n",
" '__url__': '/innovation_cache/merged_eeg/pat_96002.univariate.tar',\n",
" 'index.npy': array(10428150),\n",
" '__local_path__': '/innovation_cache/merged_eeg/pat_96002.univariate.tar',\n",
" 'label.npy': array(0, dtype=int32),\n",
" 'segment.npy': array([ 17.05792731, 28.37108812, 23.97548263, ..., -20.88427017,\n",
" -20.0133777 , 1.76995768])}"
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"sample"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [
{
"ename": "ModuleNotFoundError",
"evalue": "No module named 'autogluon'",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mModuleNotFoundError\u001b[0m Traceback (most recent call last)",
"Cell \u001b[0;32mIn[25], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mautogluon\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mtimeseries\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m TimeSeriesDataFrame, TimeSeriesPredictor\n",
"\u001b[0;31mModuleNotFoundError\u001b[0m: No module named 'autogluon'"
]
}
],
"source": [
"from autogluon.timeseries import TimeSeriesDataFrame, TimeSeriesPredictor"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [
{
"ename": "NameError",
"evalue": "name 'TimeSeriesDataFrame' is not defined",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
"Cell \u001b[0;32mIn[26], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m data \u001b[38;5;241m=\u001b[39m \u001b[43mTimeSeriesDataFrame\u001b[49m\u001b[38;5;241m.\u001b[39mfrom_path(\n\u001b[1;32m 2\u001b[0m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mhttps://autogluon.s3.amazonaws.com/datasets/timeseries/australian_electricity_subset/test.csv\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m 3\u001b[0m )\n\u001b[1;32m 4\u001b[0m data\u001b[38;5;241m.\u001b[39mhead()\n",
"\u001b[0;31mNameError\u001b[0m: name 'TimeSeriesDataFrame' is not defined"
]
}
],
"source": [
"data = TimeSeriesDataFrame.from_path(\n",
" \"https://autogluon.s3.amazonaws.com/datasets/timeseries/australian_electricity_subset/test.csv\"\n",
")\n",
"data.head()"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {},
"outputs": [
{
"ename": "NameError",
"evalue": "name 'data' is not defined",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
"Cell \u001b[0;32mIn[27], line 2\u001b[0m\n\u001b[1;32m 1\u001b[0m prediction_length \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m48\u001b[39m\n\u001b[0;32m----> 2\u001b[0m train_data, test_data \u001b[38;5;241m=\u001b[39m \u001b[43mdata\u001b[49m\u001b[38;5;241m.\u001b[39mtrain_test_split(prediction_length)\n\u001b[1;32m 4\u001b[0m predictor \u001b[38;5;241m=\u001b[39m TimeSeriesPredictor(prediction_length\u001b[38;5;241m=\u001b[39mprediction_length)\u001b[38;5;241m.\u001b[39mfit(\n\u001b[1;32m 5\u001b[0m train_data, presets\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mbolt_small\u001b[39m\u001b[38;5;124m\"\u001b[39m,\n\u001b[1;32m 6\u001b[0m )\n",
"\u001b[0;31mNameError\u001b[0m: name 'data' is not defined"
]
}
],
"source": [
"prediction_length = 48\n",
"train_data, test_data = data.train_test_split(prediction_length)\n",
"\n",
"predictor = TimeSeriesPredictor(prediction_length=prediction_length).fit(\n",
" train_data, presets=\"bolt_small\",\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {},
"outputs": [
{
"ename": "NameError",
"evalue": "name 'predictor' is not defined",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
"Cell \u001b[0;32mIn[28], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m predictions \u001b[38;5;241m=\u001b[39m \u001b[43mpredictor\u001b[49m\u001b[38;5;241m.\u001b[39mpredict(train_data)\n\u001b[1;32m 2\u001b[0m predictor\u001b[38;5;241m.\u001b[39mplot(\n\u001b[1;32m 3\u001b[0m data\u001b[38;5;241m=\u001b[39mdata,\n\u001b[1;32m 4\u001b[0m predictions\u001b[38;5;241m=\u001b[39mpredictions,\n\u001b[1;32m 5\u001b[0m item_ids\u001b[38;5;241m=\u001b[39mdata\u001b[38;5;241m.\u001b[39mitem_ids[:\u001b[38;5;241m2\u001b[39m],\n\u001b[1;32m 6\u001b[0m max_history_length\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m200\u001b[39m,\n\u001b[1;32m 7\u001b[0m );\n",
"\u001b[0;31mNameError\u001b[0m: name 'predictor' is not defined"
]
}
],
"source": [
"predictions = predictor.predict(train_data)\n",
"predictor.plot(\n",
" data=data,\n",
" predictions=predictions,\n",
" item_ids=data.item_ids[:2],\n",
" max_history_length=200,\n",
");"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Finetuning"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {},
"outputs": [
{
"ename": "NameError",
"evalue": "name 'TimeSeriesPredictor' is not defined",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
"Cell \u001b[0;32mIn[29], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m predictor \u001b[38;5;241m=\u001b[39m \u001b[43mTimeSeriesPredictor\u001b[49m(prediction_length\u001b[38;5;241m=\u001b[39mprediction_length)\u001b[38;5;241m.\u001b[39mfit(\n\u001b[1;32m 2\u001b[0m train_data\u001b[38;5;241m=\u001b[39mtrain_data,\n\u001b[1;32m 3\u001b[0m hyperparameters\u001b[38;5;241m=\u001b[39m{\n\u001b[1;32m 4\u001b[0m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mChronos\u001b[39m\u001b[38;5;124m\"\u001b[39m: [\n\u001b[1;32m 5\u001b[0m {\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodel_path\u001b[39m\u001b[38;5;124m\"\u001b[39m: \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mbolt_small\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mag_args\u001b[39m\u001b[38;5;124m\"\u001b[39m: {\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mname_suffix\u001b[39m\u001b[38;5;124m\"\u001b[39m: \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mZeroShot\u001b[39m\u001b[38;5;124m\"\u001b[39m}},\n\u001b[1;32m 6\u001b[0m {\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodel_path\u001b[39m\u001b[38;5;124m\"\u001b[39m: \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mbolt_small\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mfine_tune\u001b[39m\u001b[38;5;124m\"\u001b[39m: \u001b[38;5;28;01mTrue\u001b[39;00m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mag_args\u001b[39m\u001b[38;5;124m\"\u001b[39m: {\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mname_suffix\u001b[39m\u001b[38;5;124m\"\u001b[39m: \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mFineTuned\u001b[39m\u001b[38;5;124m\"\u001b[39m}},\n\u001b[1;32m 7\u001b[0m ]\n\u001b[1;32m 8\u001b[0m },\n\u001b[1;32m 9\u001b[0m time_limit\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m60\u001b[39m, \u001b[38;5;66;03m# time limit in seconds\u001b[39;00m\n\u001b[1;32m 10\u001b[0m enable_ensemble\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mFalse\u001b[39;00m,\n\u001b[1;32m 11\u001b[0m )\n",
"\u001b[0;31mNameError\u001b[0m: name 'TimeSeriesPredictor' is not defined"
]
}
],
"source": [
"predictor = TimeSeriesPredictor(prediction_length=prediction_length).fit(\n",
" train_data=train_data,\n",
" hyperparameters={\n",
" \"Chronos\": [\n",
" {\"model_path\": \"bolt_small\", \"ag_args\": {\"name_suffix\": \"ZeroShot\"}},\n",
" {\"model_path\": \"bolt_small\", \"fine_tune\": True, \"ag_args\": {\"name_suffix\": \"FineTuned\"}},\n",
" ]\n",
" },\n",
" time_limit=60, # time limit in seconds\n",
" enable_ensemble=False,\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {},
"outputs": [
{
"ename": "NameError",
"evalue": "name 'predictor' is not defined",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
"Cell \u001b[0;32mIn[30], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[43mpredictor\u001b[49m\u001b[38;5;241m.\u001b[39mleaderboard(test_data)\n",
"\u001b[0;31mNameError\u001b[0m: name 'predictor' is not defined"
]
}
],
"source": [
"predictor.leaderboard(test_data)\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": ".venv",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.9.20"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
Loading