{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Front Page DeepExplainer MNIST Example\n",
"\n",
"A simple example showing how to explain an MNIST CNN trained using Keras with DeepExplainer."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"x_train shape: (60000, 28, 28, 1)\n",
"60000 train samples\n",
"10000 test samples\n"
]
},
{
"data": {
"text/html": [
"Model: \"sequential\"\n",
"\n"
],
"text/plain": [
"\u001b[1mModel: \"sequential\"\u001b[0m\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
"┃ Layer (type) ┃ Output Shape ┃ Param # ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
"│ conv2d (Conv2D) │ (None, 26, 26, 32) │ 320 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ max_pooling2d (MaxPooling2D) │ (None, 13, 13, 32) │ 0 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ conv2d_1 (Conv2D) │ (None, 11, 11, 64) │ 18,496 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ max_pooling2d_1 (MaxPooling2D) │ (None, 5, 5, 64) │ 0 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ flatten (Flatten) │ (None, 1600) │ 0 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ dropout (Dropout) │ (None, 1600) │ 0 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ dense (Dense) │ (None, 10) │ 16,010 │\n",
"└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
"│ conv2d (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m26\u001b[0m, \u001b[38;5;34m26\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m320\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ max_pooling2d (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m13\u001b[0m, \u001b[38;5;34m13\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m18,496\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ max_pooling2d_1 (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ flatten (\u001b[38;5;33mFlatten\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1600\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ dropout (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1600\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m) │ \u001b[38;5;34m16,010\u001b[0m │\n",
"└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" Total params: 34,826 (136.04 KB)\n",
"\n"
],
"text/plain": [
"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m34,826\u001b[0m (136.04 KB)\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" Trainable params: 34,826 (136.04 KB)\n",
"\n"
],
"text/plain": [
"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m34,826\u001b[0m (136.04 KB)\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" Non-trainable params: 0 (0.00 B)\n",
"\n"
],
"text/plain": [
"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/3\n",
"\u001b[1m422/422\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m18s\u001b[0m 38ms/step - accuracy: 0.7699 - loss: 0.7657 - val_accuracy: 0.9788 - val_loss: 0.0789\n",
"Epoch 2/3\n",
"\u001b[1m422/422\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m26s\u001b[0m 61ms/step - accuracy: 0.9616 - loss: 0.1196 - val_accuracy: 0.9853 - val_loss: 0.0572\n",
"Epoch 3/3\n",
"\u001b[1m422/422\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 29ms/step - accuracy: 0.9737 - loss: 0.0857 - val_accuracy: 0.9862 - val_loss: 0.0492\n",
"Test loss: 0.047078389674425125\n",
"Test accuracy: 0.9847000241279602\n"
]
}
],
"source": [
"# this is the code from here --> https://github.com/keras-team/keras/blob/master/examples/demo_mnist_convnet.py\n",
"import keras\n",
"import numpy as np\n",
"from keras import layers\n",
"from keras.utils import to_categorical\n",
"\n",
"import shap\n",
"\n",
"# Model / data parameters\n",
"num_classes = 10\n",
"input_shape = (28, 28, 1)\n",
"\n",
"# Load the data and split it between train and test sets\n",
"(x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data()\n",
"\n",
"# Scale images to the [0, 1] range\n",
"x_train = x_train.astype(\"float32\") / 255\n",
"x_test = x_test.astype(\"float32\") / 255\n",
"# Make sure images have shape (28, 28, 1)\n",
"x_train = np.expand_dims(x_train, -1)\n",
"x_test = np.expand_dims(x_test, -1)\n",
"print(\"x_train shape:\", x_train.shape)\n",
"print(x_train.shape[0], \"train samples\")\n",
"print(x_test.shape[0], \"test samples\")\n",
"\n",
"\n",
"# convert class vectors to binary class matrices\n",
"y_train = to_categorical(y_train, num_classes)\n",
"y_test = to_categorical(y_test, num_classes)\n",
"\n",
"batch_size = 128\n",
"epochs = 3\n",
"\n",
"model = keras.Sequential(\n",
" [\n",
" layers.Input(shape=input_shape),\n",
" layers.Conv2D(32, kernel_size=(3, 3), activation=\"relu\"),\n",
" layers.MaxPooling2D(pool_size=(2, 2)),\n",
" layers.Conv2D(64, kernel_size=(3, 3), activation=\"relu\"),\n",
" layers.MaxPooling2D(pool_size=(2, 2)),\n",
" layers.Flatten(),\n",
" layers.Dropout(0.5),\n",
" layers.Dense(num_classes, activation=\"softmax\"),\n",
" ]\n",
")\n",
"\n",
"model.summary()\n",
"\n",
"model.compile(loss=\"categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"])\n",
"\n",
"model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs, validation_split=0.1)\n",
"\n",
"score = model.evaluate(x_test, y_test, verbose=0)\n",
"print(\"Test loss:\", score[0])\n",
"print(\"Test accuracy:\", score[1])"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"C:\\Users\\Tobias Pitters\\programming\\shap\\shap\\explainers\\_deep\\deep_tf.py:99: UserWarning: Your TensorFlow version is newer than 2.4.0 and so graph support has been removed in eager mode and some static graphs may not be supported. See PR #1483 for discussion.\n",
" warnings.warn(\"Your TensorFlow version is newer than 2.4.0 and so graph support has been removed in eager mode and some static graphs may not be supported. See PR #1483 for discussion.\")\n"
]
},
{
"ename": "AttributeError",
"evalue": "'tuple' object has no attribute 'as_list'",
"output_type": "error",
"traceback": [
"\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[1;31mAttributeError\u001b[0m Traceback (most recent call last)",
"Cell \u001b[1;32mIn[2], line 5\u001b[0m\n\u001b[0;32m 2\u001b[0m background \u001b[38;5;241m=\u001b[39m x_train[np\u001b[38;5;241m.\u001b[39mrandom\u001b[38;5;241m.\u001b[39mchoice(x_train\u001b[38;5;241m.\u001b[39mshape[\u001b[38;5;241m0\u001b[39m], \u001b[38;5;241m100\u001b[39m, replace\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mFalse\u001b[39;00m)]\n\u001b[0;32m 4\u001b[0m \u001b[38;5;66;03m# explain predictions of the model on three images\u001b[39;00m\n\u001b[1;32m----> 5\u001b[0m e \u001b[38;5;241m=\u001b[39m \u001b[43mshap\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mDeepExplainer\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mbackground\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m 6\u001b[0m \u001b[38;5;66;03m# ...or pass tensors directly\u001b[39;00m\n\u001b[0;32m 7\u001b[0m \u001b[38;5;66;03m# e = shap.DeepExplainer((model.layers[0].input, model.layers[-1].output), background)\u001b[39;00m\n\u001b[0;32m 8\u001b[0m shap_values \u001b[38;5;241m=\u001b[39m e\u001b[38;5;241m.\u001b[39mshap_values(x_test[\u001b[38;5;241m0\u001b[39m:\u001b[38;5;241m5\u001b[39m])\n",
"File \u001b[1;32m~\\programming\\shap\\shap\\explainers\\_deep\\__init__.py:90\u001b[0m, in \u001b[0;36mDeepExplainer.__init__\u001b[1;34m(self, model, data, session, learning_phase_flags)\u001b[0m\n\u001b[0;32m 87\u001b[0m \u001b[38;5;28msuper\u001b[39m()\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__init__\u001b[39m(model, masker)\n\u001b[0;32m 89\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m framework \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtensorflow\u001b[39m\u001b[38;5;124m'\u001b[39m:\n\u001b[1;32m---> 90\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mexplainer \u001b[38;5;241m=\u001b[39m \u001b[43mTFDeep\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdata\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43msession\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mlearning_phase_flags\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m 91\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m framework \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mpytorch\u001b[39m\u001b[38;5;124m'\u001b[39m:\n\u001b[0;32m 92\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mexplainer \u001b[38;5;241m=\u001b[39m PyTorchDeep(model, data)\n",
"File \u001b[1;32m~\\programming\\shap\\shap\\explainers\\_deep\\deep_tf.py:172\u001b[0m, in \u001b[0;36mTFDeep.__init__\u001b[1;34m(self, model, data, session, learning_phase_flags)\u001b[0m\n\u001b[0;32m 170\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mphi_symbolics \u001b[38;5;241m=\u001b[39m [\u001b[38;5;28;01mNone\u001b[39;00m]\n\u001b[0;32m 171\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m--> 172\u001b[0m noutputs \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmodel_output\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mshape\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mas_list\u001b[49m()[\u001b[38;5;241m1\u001b[39m]\n\u001b[0;32m 173\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m noutputs \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m 174\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mphi_symbolics \u001b[38;5;241m=\u001b[39m [\u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;28;01mfor\u001b[39;00m i \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(noutputs)]\n",
"\u001b[1;31mAttributeError\u001b[0m: 'tuple' object has no attribute 'as_list'"
]
}
],
"source": [
"# select a set of background examples to take an expectation over\n",
"background = x_train[np.random.choice(x_train.shape[0], 100, replace=False)]\n",
"\n",
"# explain predictions of the model on three images\n",
"e = shap.DeepExplainer(model, background)\n",
"# ...or pass tensors directly\n",
"# e = shap.DeepExplainer((model.layers[0].input, model.layers[-1].output), background)\n",
"shap_values = e.shap_values(x_test[0:5])"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# plot the feature attributions\n",
"shap.image_plot(shap_values, -x_test[0:5])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The plot above shows the explanations for each class on five predictions. Note that the explanations are ordered for the classes 0-9 going left to right along the rows."
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"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": "3.11.6"
}
},
"nbformat": 4,
"nbformat_minor": 2
}