Skip to content

Commit

Permalink
docs(notebook): fresh run²
Browse files Browse the repository at this point in the history
  • Loading branch information
@marcpinet
marcpinet committed Dec 6, 2024
1 parent 3211694 commit bfbe103
Showing 1 changed file with 31 additions and 56 deletions.
87 changes: 31 additions & 56 deletions examples/models-usages/mlp-classification-regression/sentiment_analysis.ipynb
View file
Original file line number Diff line number Diff line change
Expand Up @@ -68,7 +68,7 @@
},
{
"cell_type": "code",
"execution_count": null,
"execution_count": 3,
"metadata": {
"ExecuteTime": {
"end_time": "2024-11-14T19:16:46.274852700Z",
Expand All @@ -84,35 +84,21 @@
"x_test shape: (25000, 200)\n",
"y_train shape: (20000,)\n",
"y_test shape: (25000,)\n",
"x_train[0]: [4.500e+01 1.080e+02 1.000e+01 1.000e+01 1.100e+01 4.000e+00 6.500e+01\n",
" 3.960e+03 9.000e+00 1.100e+01 4.100e+01 4.020e+02 2.000e+00 7.800e+02\n",
" 3.300e+01 2.000e+00 6.130e+03 1.100e+01 2.000e+00 4.000e+00 2.763e+03\n",
" 8.440e+02 2.600e+01 2.000e+00 2.240e+02 5.000e+00 1.930e+02 3.960e+03\n",
" 3.900e+01 4.400e+01 7.900e+02 1.530e+02 1.540e+02 1.430e+02 4.100e+01\n",
" 2.521e+03 5.600e+01 8.000e+00 4.100e+01 2.028e+03 5.590e+02 1.100e+01\n",
" 4.000e+00 2.000e+01 4.400e+01 6.383e+03 5.284e+03 4.740e+02 4.820e+02\n",
" 1.300e+01 6.600e+01 9.200e+01 1.040e+02 2.250e+02 6.000e+00 4.040e+02\n",
" 5.240e+02 1.800e+01 3.960e+03 1.800e+01 1.110e+02 7.000e+00 1.780e+02\n",
" 3.960e+03 4.510e+02 4.420e+02 7.600e+01 9.900e+01 9.760e+02 6.000e+00\n",
" 1.369e+03 1.100e+01 2.630e+02 2.000e+00 4.600e+02 8.519e+03 2.000e+00\n",
" 9.000e+00 3.084e+03 5.900e+01 9.000e+00 5.500e+01 7.207e+03 2.000e+00\n",
" 5.000e+00 2.000e+00 5.900e+01 4.700e+01 7.750e+02 7.000e+00 9.963e+03\n",
" 5.900e+01 4.700e+01 6.000e+00 8.700e+01 3.930e+02 3.100e+01 1.500e+01\n",
" 3.775e+03 1.100e+01 1.290e+02 3.300e+02 7.300e+01 1.030e+02 4.000e+00\n",
" 2.000e+01 9.000e+00 1.200e+02 1.793e+03 8.000e+00 2.000e+00 2.000e+00\n",
" 2.000e+00 5.071e+03 3.960e+03 4.700e+01 2.470e+02 6.000e+00 5.879e+03\n",
" 8.220e+02 7.400e+01 2.000e+00 2.100e+01 1.460e+02 1.688e+03 8.000e+00\n",
" 4.909e+03 1.500e+01 4.800e+01 2.000e+00 1.999e+03 1.100e+01 4.000e+00\n",
" 2.170e+02 1.300e+01 1.040e+02 5.900e+01 8.000e+01 2.700e+03 8.300e+01\n",
" 1.200e+01 4.300e+01 1.700e+01 3.960e+03 3.418e+03 5.300e+01 9.760e+02\n",
" 5.000e+00 6.861e+03 1.700e+01 5.900e+01 2.140e+02 9.220e+02 2.000e+00\n",
" 4.600e+02 5.603e+03 2.000e+00 4.860e+02 5.000e+00 1.557e+03 2.000e+00\n",
" 5.500e+01 7.300e+01 1.400e+02 1.404e+03 5.000e+00 8.510e+02 1.400e+01\n",
" 2.000e+01 4.500e+01 2.400e+01 4.000e+01 2.330e+02 3.340e+02 8.740e+02\n",
" 1.100e+02 5.000e+00 6.000e+01 1.510e+02 1.200e+01 1.600e+01 5.260e+02\n",
" 3.400e+01 1.091e+03 2.000e+00 1.200e+01 9.000e+00 3.680e+02 7.000e+00\n",
" 2.000e+00 2.442e+03 8.700e+01 3.700e+02 1.102e+03 1.524e+03 5.000e+00\n",
" 7.300e+01 2.240e+02 2.060e+02 8.440e+02]\n",
"x_train[0]: [ 45 108 10 10 11 4 65 3960 9 11 41 402 2 780\n",
" 33 2 6130 11 2 4 2763 844 26 2 224 5 193 3960\n",
" 39 44 790 153 154 143 41 2521 56 8 41 2028 559 11\n",
" 4 20 44 6383 5284 474 482 13 66 92 104 225 6 404\n",
" 524 18 3960 18 111 7 178 3960 451 442 76 99 976 6\n",
" 1369 11 263 2 460 8519 2 9 3084 59 9 55 7207 2\n",
" 5 2 59 47 775 7 9963 59 47 6 87 393 31 15\n",
" 3775 11 129 330 73 103 4 20 9 120 1793 8 2 2\n",
" 2 5071 3960 47 247 6 5879 822 74 2 21 146 1688 8\n",
" 4909 15 48 2 1999 11 4 217 13 104 59 80 2700 83\n",
" 12 43 17 3960 3418 53 976 5 6861 17 59 214 922 2\n",
" 460 5603 2 486 5 1557 2 55 73 140 1404 5 851 14\n",
" 20 45 24 40 233 334 874 110 5 60 151 12 16 526\n",
" 34 1091 2 12 9 368 7 2 2442 87 370 1102 1524 5\n",
" 73 224 206 844]\n",
"y_train[0]: 1\n"
]
}
Expand Down Expand Up @@ -184,18 +170,19 @@
"name": "stdout",
"output_type": "stream",
"text": [
"Sequential(temperature=1.0, gradient_clip_threshold=5.0, enable_padding=False, padding_size=32, random_state=1731611806261338000)\n",
"Sequential(gradient_clip_threshold=5.0, enable_padding=False, padding_size=32, random_state=1733520050429276600)\n",
"-------------------------------------------------\n",
"Layer 1: Input(input_shape=(200,))\n",
"Layer 2: Embedding(input_dim=10000, output_dim=100)\n",
"Layer 3: Bidirectional(layer=LSTM(units=32, return_sequences=True, return_state=False, random_state=None, clip_value=5.0))\n",
"Layer 3: Bidirectional(layer=LSTM(units=32, return_sequences=True, return_state=False, clip_value=5.0, random_state=None))\n",
"Layer 4: Attention(use_scale=True, score_mode=dot, return_sequences=False)\n",
"Layer 5: Dense(units=1)\n",
"Layer 6: Activation(Sigmoid)\n",
"-------------------------------------------------\n",
"Loss function: BinaryCrossentropy\n",
"Optimizer: Adam(learning_rate=0.001, beta_1=0.9, beta_2=0.999, epsilon=1e-08, clip_norm=None, clip_value=None)\n",
"-------------------------------------------------\n"
"-------------------------------------------------\n",
"\n"
]
}
],
Expand All @@ -214,7 +201,7 @@
},
{
"cell_type": "code",
"execution_count": 6,
"execution_count": null,
"metadata": {
"ExecuteTime": {
"end_time": "2024-11-14T19:49:32.469776300Z",
Expand All @@ -226,29 +213,17 @@
"name": "stdout",
"output_type": "stream",
"text": [
"[==============================] 100% Epoch 1/10 - loss: 0.4515 - accuracy: 0.8001 - 149.41s - val_accuracy: 0.8397\n",
"[==============================] 100% Epoch 2/10 - loss: 0.2670 - accuracy: 0.8926 - 134.93s - val_accuracy: 0.8542\n",
"[==============================] 100% Epoch 3/10 - loss: 0.2317 - accuracy: 0.9129 - 136.07s - val_accuracy: 0.8512\n",
"[==============================] 100% Epoch 4/10 - loss: 0.2437 - accuracy: 0.9196 - 133.42s - val_accuracy: 0.8383\n",
"[==============================] 100% Epoch 5/10 - loss: 0.2506 - accuracy: 0.9280 - 135.57s - val_accuracy: 0.8449\n",
"[==============================] 100% Epoch 6/10 - loss: 0.2753 - accuracy: 0.9333 - 138.26s - val_accuracy: 0.8346\n",
"[==============================] 100% Epoch 7/10 - loss: 0.3047 - accuracy: 0.9371 - 141.18s - val_accuracy: 0.8236\n",
"[==============================] 100% Epoch 8/10 - loss: 0.3261 - accuracy: 0.9405 - 140.46s - val_accuracy: 0.8178\n",
"[==============================] 100% Epoch 9/10 - loss: 0.3593 - accuracy: 0.9459 - 135.74s - val_accuracy: 0.8236\n",
"[==============================] 100% Epoch 10/10 - loss: 0.3402 - accuracy: 0.9528 - 144.90s - val_accuracy: 0.8296\n"
"[==============================] 100% Epoch 1/5 - 274.36s - loss: 0.6359 - accuracy: 0.6967 - val_loss: 0.7389 - val_accuracy: 0.7740\n",
"[==============================] 100% Epoch 2/5 - 276.99s - loss: 0.4441 - accuracy: 0.8237 - val_loss: 1.0205 - val_accuracy: 0.8307\n",
"[==============================] 100% Epoch 3/5 - 285.17s - loss: 0.3278 - accuracy: 0.8611 - val_loss: 1.4672 - val_accuracy: 0.8485\n",
"[==============================] 100% Epoch 4/5 - 269.14s - loss: 0.2853 - accuracy: 0.8797 - val_loss: 1.9860 - val_accuracy: 0.8568\n",
"[==============================] 100% Epoch 5/5 - 267.17s - loss: 0.2713 - accuracy: 0.8895 - val_loss: 2.5888 - val_accuracy: 0.8598\n",
"\n"
]
},
{
"data": {
"text/plain": []
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"model.fit(x_train, y_train, epochs=10, batch_size=32, validation_data=(x_test, y_test), metrics=['accuracy'], random_state=42)"
"model.fit(x_train, y_train, epochs=5, batch_size=32, validation_data=(x_test, y_test), metrics=['accuracy'], random_state=42)"
]
},
{
Expand All @@ -260,7 +235,7 @@
},
{
"cell_type": "code",
"execution_count": 7,
"execution_count": 9,
"metadata": {
"ExecuteTime": {
"end_time": "2024-11-14T19:49:59.259717400Z",
Expand All @@ -272,8 +247,8 @@
"name": "stdout",
"output_type": "stream",
"text": [
"Loss: 11.780504039134605\n",
"Accuracy: 0.831\n"
"Loss: 2.6114417790014\n",
"Accuracy: 0.8712\n"
]
}
],
Expand Down

0 comments on commit bfbe103

Please sign in to comment.