6-Black_box-VS-White_box_models-Interpretations-Surrogate_models.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[],"toc_visible":true,"authorship_tag":"ABX9TyOP89hK/YThSQJl0lZKgFUr"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"},"widgets":{"application/vnd.jupyter.widget-state+json":{"0841b7b341404a8db5211df303da465a":{"model_module":"@jupyter-widgets/controls","model_name":"VBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":["widget-interact"],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"VBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"VBoxView","box_style":"","children":["IPY_MODEL_a16628b03ea946208bcb7545e1e2fee7","IPY_MODEL_37c7a1a67e0c4ae29b283815bd9b0a3f"],"layout":"IPY_MODEL_26992448d82942fe9a3174a6f2000e47"}},"a16628b03ea946208bcb7545e1e2fee7":{"model_module":"@jupyter-widgets/controls","model_name":"IntSliderModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"IntSliderModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"IntSliderView","continuous_update":true,"description":"instance","description_tooltip":null,"disabled":false,"layout":"IPY_MODEL_bbc320c5df764d089a3cfb640377559f","max":9,"min":0,"orientation":"horizontal","readout":true,"readout_format":"d","step":1,"style":"IPY_MODEL_e44e5745056d46fd8232436cc79c4c84","value":0}},"37c7a1a67e0c4ae29b283815bd9b0a3f":{"model_module":"@jupyter-widgets/output","model_name":"OutputModel","model_module_version":"1.0.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/output","_model_module_version":"1.0.0","_model_name":"OutputModel","_view_count":null,"_view_module":"@jupyter-widgets/output","_view_module_version":"1.0.0","_view_name":"OutputView","layout":"IPY_MODEL_86738edd5a11444886fc595679341828","msg_id":"","outputs":[{"output_type":"stream","name":"stdout","text":["Real value: 321.0\n","Predicted value: 214.59858063690152\n","Sum(weights*instance): 0.07223821372377162 + Intercept (Bias): -0.01250504296931576 = 0.05973317075445586 > 0 -> 1\n","Number of features: 10\n"]},{"output_type":"display_data","data":{"text/plain":"<Figure size 800x600 with 1 Axes>","image/png":"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\n"},"metadata":{}}]}},"26992448d82942fe9a3174a6f2000e47":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"bbc320c5df764d089a3cfb640377559f":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"e44e5745056d46fd8232436cc79c4c84":{"model_module":"@jupyter-widgets/controls","model_name":"SliderStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"SliderStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":"","handle_color":null}},"86738edd5a11444886fc595679341828":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"15f701fb400a4f1ab093bdf7a0aceae6":{"model_module":"@jupyter-widgets/controls","model_name":"VBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":["widget-interact"],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"VBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"VBoxView","box_style":"","children":["IPY_MODEL_876d4ae0092f446c977dc360865123c5","IPY_MODEL_db703ce14cb8494fa941e9fea3a0bdde"],"layout":"IPY_MODEL_64291ffe1f0e4c738f80958f2a2456d8"}},"876d4ae0092f446c977dc360865123c5":{"model_module":"@jupyter-widgets/controls","model_name":"IntSliderModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"IntSliderModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"IntSliderView","continuous_update":true,"description":"instance","description_tooltip":null,"disabled":false,"layout":"IPY_MODEL_1987362c389444a4a4a2d95b44c712d2","max":9,"min":0,"orientation":"horizontal","readout":true,"readout_format":"d","step":1,"style":"IPY_MODEL_efc4585a93ba482ba1f230b6e0753676","value":0}},"db703ce14cb8494fa941e9fea3a0bdde":{"model_module":"@jupyter-widgets/output","model_name":"OutputModel","model_module_version":"1.0.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/output","_model_module_version":"1.0.0","_model_name":"OutputModel","_view_count":null,"_view_module":"@jupyter-widgets/output","_view_module_version":"1.0.0","_view_name":"OutputView","layout":"IPY_MODEL_141f3869f0d54641b0b2855263060d22","msg_id":"","outputs":[{"output_type":"stream","name":"stdout","text":["Real value: 321.0\n","Predicted value: 232.51202752156635\n","Sum(weights*instance): 0.07223821372377162 + Intercept (Bias): -0.01250504296931576 = 0.05973317075445586 > 0 -> 1\n","Number of features: 10\n"]},{"output_type":"display_data","data":{"text/plain":"<Figure size 800x600 with 1 Axes>","image/png":"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\n"},"metadata":{}}]}},"64291ffe1f0e4c738f80958f2a2456d8":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"1987362c389444a4a4a2d95b44c712d2":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"efc4585a93ba482ba1f230b6e0753676":{"model_module":"@jupyter-widgets/controls","model_name":"SliderStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"SliderStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":"","handle_color":null}},"141f3869f0d54641b0b2855263060d22":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}}}}},"cells":[{"cell_type":"markdown","source":["# Imports"],"metadata":{"id":"LGiu8xjwmqdK"}},{"cell_type":"code","source":["import numpy as np\n","import pandas as pd\n","import matplotlib.pyplot as plt\n","import seaborn as sns\n","\n","from sklearn.datasets import load_diabetes\n","from sklearn.model_selection import train_test_split\n","from sklearn.preprocessing import MinMaxScaler\n","from sklearn.metrics import mean_squared_error as mse\n","from sklearn.metrics import mean_absolute_error as mae\n","from sklearn.metrics import mean_absolute_percentage_error as mape\n","from sklearn.linear_model import LinearRegression\n","from sklearn.neural_network import MLPRegressor\n","from sklearn.neighbors import KNeighborsRegressor\n","\n","from IPython.display import SVG\n","from IPython.display import display \n","from ipywidgets import interactive"],"metadata":{"id":"OHdwFE1Vmryi","executionInfo":{"status":"ok","timestamp":1686337706974,"user_tz":-180,"elapsed":8,"user":{"displayName":"Achil Pal","userId":"04608072963754583562"}}},"execution_count":53,"outputs":[]},{"cell_type":"markdown","source":["# Helpers"],"metadata":{"id":"R0hVV9pjmsRy"}},{"cell_type":"code","source":["def plot_sensor(instance=0):\n","  # Get instance\n","  random_instance = X_test[instance]\n","\n","  # Print real and true values\n","  print(\"Real value:\", y_test[instance])\n","  print(\"Predicted value:\", y_test_preds[instance])\n","\n","  # Get \"weights\", \"bias\" and \"summation\" variables\n","  weights = model.coef_\n","  summation = sum(weights * random_instance)\n","  bias = model.intercept_\n","\n","  # Get and print residual\n","  res = \"\"\n","  if (summation + bias > 0):\n","    res = \" > 0 -> 1\"\n","  else:\n","    res = \" <= 0 -> 0\"\n","  print(\"Sum(weights*instance): \"+str(summation)+\" + Intercept (Bias): \"+str(bias)+\" = \"+ str(summation+bias)+ res)\n","\n","  # Sort weights (coefficients) by absolute value\n","  model_weights = pd.DataFrame({'features': list(feature_names),\n","                                'weights*values': list(weights * random_instance)})\n","  model_weights = model_weights.reindex(model_weights['weights*values'].abs().sort_values(ascending=False).index)\n","  model_weights = model_weights[(model_weights[\"weights*values\"] != 0)]    \n","\n","  # print number of features\n","  print(\"Number of features:\",len(model_weights.values))\n","\n","  # plot weights' importance as a bar plot\n","  plt.figure(num=None, figsize=(8, 6), dpi=100, facecolor='w', edgecolor='k')\n","  sns.barplot(x=\"weights*values\", y=\"features\", data=model_weights)\n","  plt.xticks(rotation=90)\n","  plt.show()"],"metadata":{"id":"K2PCzwbTmthh","executionInfo":{"status":"ok","timestamp":1686337706975,"user_tz":-180,"elapsed":8,"user":{"displayName":"Achil Pal","userId":"04608072963754583562"}}},"execution_count":54,"outputs":[]},{"cell_type":"markdown","source":["# Get and Preprocess the data"],"metadata":{"id":"3FvneEeLn644"}},{"cell_type":"code","source":["# Get the data\n","diabetes = load_diabetes()\n","X = diabetes.data\n","y = diabetes.target\n","feature_names = diabetes.feature_names\n","\n","# Split the data into train and test sets\n","X_train, X_test, y_train, y_test = train_test_split(X, y, \n","                                                    test_size=0.3,\n","                                                    random_state=0, \n","                                                    shuffle=True)\n","\n","# Target position\n","TARGET_POS = -1   \n","\n","# Concatenate data, so as to scale both features and the target\n","train_data = np.concatenate((X_train, y_train.reshape(-1, 1)), axis=1)\n","test_data = np.concatenate((X_test, y_test.reshape(-1, 1)), axis=1)\n","\n","# Scale the data\n","scaler = MinMaxScaler()\n","train_data_sc = scaler.fit_transform(train_data)\n","test_data_sc = scaler.transform(test_data)\n","\n","# Get scaled features and the scaled target\n","X_train_sc = train_data_sc[:, :TARGET_POS]\n","y_train_sc = train_data_sc[:, TARGET_POS]\n","X_test_sc = test_data_sc[:, :TARGET_POS]\n","y_test_sc = test_data_sc[:, TARGET_POS]\n","\n","print(X_train_sc.shape, X_test_sc.shape, y_train_sc.shape, y_test_sc.shape)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"KWGuNvXMn928","executionInfo":{"status":"ok","timestamp":1686337706975,"user_tz":-180,"elapsed":8,"user":{"displayName":"Achil Pal","userId":"04608072963754583562"}},"outputId":"fa50dc58-cd9e-4725-8a45-edf83a8d488b"},"execution_count":55,"outputs":[{"output_type":"stream","name":"stdout","text":["(309, 10) (133, 10) (309,) (133,)\n"]}]},{"cell_type":"markdown","source":["# 1. White Box model"],"metadata":{"id":"N-MCjbynqKYq"}},{"cell_type":"markdown","source":["## Train a linear regression model"],"metadata":{"id":"Y2DyTCM8qTYr"}},{"cell_type":"code","source":["# Initialize and train linear model\n","model = LinearRegression()\n","model.fit(X_train_sc, y_train_sc)\n","\n","# Get scaled predictions\n","y_train_preds_sc = model.predict(X_train_sc)\n","y_test_preds_sc = model.predict(X_test_sc)\n","\n","# Unscale predictions\n","y_train_preds = scaler.min_[TARGET_POS] + y_train_preds_sc * scaler.data_range_[TARGET_POS]\n","y_test_preds = scaler.min_[TARGET_POS] + y_test_preds_sc * scaler.data_range_[TARGET_POS]\n","\n","# Print results (MAE, MSE, MAPE)\n","print(f\"***** Linear Regression Model Performance *****\")\n","print(f\"Train set: MAE = {mae(y_train, y_train_preds)}, MSE = {mse(y_train, y_train_preds)}, MAPE = {100 * mape(y_train, y_train_preds)}%\")\n","print(f\"Test set: MAE = {mae(y_test, y_test_preds)}, MSE = {mse(y_test, y_test_preds)}, MAPE = {100 * mape(y_test, y_test_preds)}%\")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Qmjt3w09qL4o","executionInfo":{"status":"ok","timestamp":1686337706975,"user_tz":-180,"elapsed":6,"user":{"displayName":"Achil Pal","userId":"04608072963754583562"}},"outputId":"4e1e5197-77b5-47f7-b20c-f34e2f2c2301"},"execution_count":56,"outputs":[{"output_type":"stream","name":"stdout","text":["***** Linear Regression Model Performance *****\n","Train set: MAE = 47.31257664821065, MSE = 3433.0437075884915, MAPE = 36.325356417456554%\n","Test set: MAE = 46.12417364230234, MSE = 3461.7379665593207, MAPE = 35.170255858383044%\n"]}]},{"cell_type":"markdown","source":["## Global interpretation: plot the weights (coefficient of the features = feature importance) assigned by the model to each feature"],"metadata":{"id":"5oM6hQwMsjAm"}},{"cell_type":"code","source":["# Get weights\n","weights = model.coef_\n","model_weights = pd.DataFrame({'features': list(feature_names), \n","                              'weights': list(weights)})\n","\n","# Sort weights (coefficients) by absolute value\n","model_weights = model_weights.reindex(model_weights['weights'].abs().sort_values(ascending=False).index) \n","model_weights = model_weights[(model_weights[\"weights\"] != 0)]   \n","\n","print(\"Number of features:\",len(model_weights.values))\n","\n","# plot weights' importance as a bar plot\n","plt.figure(num=None, figsize=(8, 6), dpi=100, facecolor='w', edgecolor='k')\n","sns.barplot(x=\"weights\", y=\"features\", data=model_weights)\n","plt.title(\"Intercept (Bias): \" + str(model.intercept_), loc='right')\n","plt.xticks(rotation=90)\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":0},"id":"gCSN4kruswUo","executionInfo":{"status":"ok","timestamp":1686337707291,"user_tz":-180,"elapsed":321,"user":{"displayName":"Achil Pal","userId":"04608072963754583562"}},"outputId":"2e1249ce-fd77-4e4b-bafa-bb534e028da3"},"execution_count":57,"outputs":[{"output_type":"stream","name":"stdout","text":["Number of features: 10\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 800x600 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"markdown","source":["## Local interpetation"],"metadata":{"id":"YcUq0KVFwUyB"}},{"cell_type":"code","source":["inter = interactive(plot_sensor, instance=(0,9))\n","display(inter)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":0,"referenced_widgets":["0841b7b341404a8db5211df303da465a","a16628b03ea946208bcb7545e1e2fee7","37c7a1a67e0c4ae29b283815bd9b0a3f","26992448d82942fe9a3174a6f2000e47","bbc320c5df764d089a3cfb640377559f","e44e5745056d46fd8232436cc79c4c84","86738edd5a11444886fc595679341828"]},"id":"11dE1QOVwXlV","executionInfo":{"status":"ok","timestamp":1686337707630,"user_tz":-180,"elapsed":350,"user":{"displayName":"Achil Pal","userId":"04608072963754583562"}},"outputId":"15771a71-0dfd-4415-b1b2-3dfe87dcf1e1"},"execution_count":58,"outputs":[{"output_type":"display_data","data":{"text/plain":["interactive(children=(IntSlider(value=0, description='instance', max=9), Output()), _dom_classes=('widget-inte…"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"0841b7b341404a8db5211df303da465a"}},"metadata":{}}]},{"cell_type":"markdown","source":["# 2. Black Box model"],"metadata":{"id":"7TjVrnTpy_2W"}},{"cell_type":"markdown","source":["## Train a Neural Network as black box model"],"metadata":{"id":"8Uhkl604zuWQ"}},{"cell_type":"code","source":["# Initialize and train model\n","UNITS_PER_LAYER = 100\n","regressor = MLPRegressor(hidden_layer_sizes=(UNITS_PER_LAYER, UNITS_PER_LAYER))\n","regressor.fit(X_train_sc, y_train_sc)\n","\n","# Get scaled predictions\n","y_test_preds_sc = regressor.predict(X_test_sc)\n","\n","# Unscale predictions\n","y_test_preds = scaler.min_[TARGET_POS] + y_test_preds_sc * scaler.data_range_[TARGET_POS]\n","\n","# Print results (MAE, MSE, MAPE)\n","print(f\"***** MLP Model Performance *****\")\n","print(f\"Test set: MAE = {mae(y_test, y_test_preds)}, MSE = {mse(y_test, y_test_preds)}, MAPE = {100 * mape(y_test, y_test_preds)}%\")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"dAkGe1PgzCmk","executionInfo":{"status":"ok","timestamp":1686337707984,"user_tz":-180,"elapsed":357,"user":{"displayName":"Achil Pal","userId":"04608072963754583562"}},"outputId":"4dd49214-c40a-4bee-9e14-036938811d5b"},"execution_count":59,"outputs":[{"output_type":"stream","name":"stdout","text":["***** MLP Model Performance *****\n","Test set: MAE = 47.12270716248503, MSE = 3664.0971552905553, MAPE = 35.26724666449434%\n"]}]},{"cell_type":"markdown","source":["## Obtain predictions of the black box on the training set (could perhaps better be done via cross-validation)"],"metadata":{"id":"yBdN9Seb0hsP"}},{"cell_type":"code","source":["# Get scaled predictions\n","y_train_preds_sc = regressor.predict(X_train_sc)\n","\n","# Unscale predictions\n","y_train_preds = scaler.min_[TARGET_POS] + y_train_preds_sc * scaler.data_range_[TARGET_POS]\n","\n","# Print results (MAE, MSE, MAPE)\n","print(f\"***** MLP Model Performance *****\")\n","print(f\"Train set: MAE = {mae(y_train, y_train_preds)}, MSE = {mse(y_train, y_train_preds)}, MAPE = {100 * mape(y_train, y_train_preds)}%\")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"hu0sZiSS0nCT","executionInfo":{"status":"ok","timestamp":1686337707985,"user_tz":-180,"elapsed":9,"user":{"displayName":"Achil Pal","userId":"04608072963754583562"}},"outputId":"565dbad1-d0e3-4fe9-da97-ea0a5ef62e87"},"execution_count":60,"outputs":[{"output_type":"stream","name":"stdout","text":["***** MLP Model Performance *****\n","Train set: MAE = 45.08969234878358, MSE = 3122.288279184051, MAPE = 34.189780112953656%\n"]}]},{"cell_type":"markdown","source":["## Global Surrogate models: explanation with a Linear model"],"metadata":{"id":"4XV0QvS63Q-J"}},{"cell_type":"code","source":["# Initialize a linear model\n","linear_model = LinearRegression()\n","\n","# new_X_train = X_train_sc (scaled version)\n","# new_y_train = y_train_preds_sc (scaled version)\n","\n","# Train the model\n","linear_model.fit(X_train_sc, y_train_preds_sc)\n","\n","# Get scaled predictions\n","y_test_linear_preds_sc = linear_model.predict(X_test_sc)\n","\n","# Unscale predictions\n","y_test_linear_preds = scaler.min_[TARGET_POS] + y_test_linear_preds_sc * scaler.data_range_[TARGET_POS]\n","\n","# Print results (MAE, MSE, MAPE)\n","print(f\"***** Simple Linear Model Performance *****\")\n","print(f\"Train set: MAE = {mae(y_test, y_test_linear_preds)}, MSE = {mse(y_test, y_test_linear_preds)}, MAPE = {100 * mape(y_test, y_test_linear_preds)}%\")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"3yYBPnUh2vZd","executionInfo":{"status":"ok","timestamp":1686337707986,"user_tz":-180,"elapsed":9,"user":{"displayName":"Achil Pal","userId":"04608072963754583562"}},"outputId":"21f6c4b0-d1a6-465e-b937-f7bae1142d8b"},"execution_count":61,"outputs":[{"output_type":"stream","name":"stdout","text":["***** Simple Linear Model Performance *****\n","Train set: MAE = 46.928391003010034, MSE = 3617.3078302844347, MAPE = 35.59782459440513%\n"]}]},{"cell_type":"code","source":["# Get weights\n","weights = model.coef_\n","model_weights = pd.DataFrame({'features': list(feature_names), \n","                              'weights': list(weights)})\n","\n","# Sort weights (coefficients) by absolute value\n","model_weights = model_weights.reindex(model_weights['weights'].abs().sort_values(ascending=False).index) \n","model_weights = model_weights[(model_weights[\"weights\"] != 0)]   \n","\n","print(\"Number of features:\",len(model_weights.values))\n","\n","# plot weights' importance as a bar plot\n","plt.figure(num=None, figsize=(8, 6), dpi=100, facecolor='w', edgecolor='k')\n","sns.barplot(x=\"weights\", y=\"features\", data=model_weights)\n","plt.title(\"Intercept (Bias): \" + str(model.intercept_), loc='center')\n","plt.xticks(rotation=90)\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":0},"id":"Qj1d_hXa4do0","executionInfo":{"status":"ok","timestamp":1686337708541,"user_tz":-180,"elapsed":564,"user":{"displayName":"Achil Pal","userId":"04608072963754583562"}},"outputId":"81a2afa0-6a62-4f51-d982-a285bc255a09"},"execution_count":62,"outputs":[{"output_type":"stream","name":"stdout","text":["Number of features: 10\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 800x600 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"markdown","source":["## Local Surrogate models: explanation with a linear model"],"metadata":{"id":"DYjiHzV77Zm-"}},{"cell_type":"code","source":["# Initialize a K-Neighbors model\n","knn_model = KNeighborsRegressor(n_neighbors=50, weights=\"distance\", metric=\"minkowski\", p=2)\n","\n","# Fit the model on the new X_train and the new y_train\n","knn_model.fit(X_train_sc, y_train_preds_sc)\n","\n","# Find neighbors\n","print(\"Finding Neighbors of Instance...\")\n","instance = X_test_sc[0]\n","test_x = [instance]\n","neighbors = knn_model.kneighbors(test_x, n_neighbors=50, return_distance=False).flatten()\n","\n","# Create new X's and y's\n","new_X_train2 = []\n","new_y_train2 = []\n","for i in neighbors:\n","  new_X_train2.append(X_train_sc[i])\n","  new_y_train2.append(y_train_preds_sc[i])"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"UYrBFltS71Em","executionInfo":{"status":"ok","timestamp":1686337708542,"user_tz":-180,"elapsed":10,"user":{"displayName":"Achil Pal","userId":"04608072963754583562"}},"outputId":"baa6e2d0-db13-451d-e5f6-a6c7e631cb56"},"execution_count":63,"outputs":[{"output_type":"stream","name":"stdout","text":["Finding Neighbors of Instance...\n"]}]},{"cell_type":"code","source":["# Initialize and train a linear model\n","linear_model = LinearRegression()\n","linear_model.fit(new_X_train2, new_y_train2)\n","\n","# Get scaled predictions\n","y_test_linear_preds_sc = linear_model.predict(X_test_sc)\n","\n","# Unscale predictions\n","y_test_linear_preds = scaler.min_[TARGET_POS] + y_test_linear_preds_sc * scaler.data_range_[TARGET_POS]\n","\n","# Display results\n","inter = interactive(plot_sensor, instance=(0,9))\n","display(inter)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":664,"referenced_widgets":["15f701fb400a4f1ab093bdf7a0aceae6","876d4ae0092f446c977dc360865123c5","db703ce14cb8494fa941e9fea3a0bdde","64291ffe1f0e4c738f80958f2a2456d8","1987362c389444a4a4a2d95b44c712d2","efc4585a93ba482ba1f230b6e0753676","141f3869f0d54641b0b2855263060d22"]},"id":"2VYXbRNW5pIW","executionInfo":{"status":"ok","timestamp":1686337708876,"user_tz":-180,"elapsed":343,"user":{"displayName":"Achil Pal","userId":"04608072963754583562"}},"outputId":"33f6867c-1279-4a18-a37d-8f5bdf88e912"},"execution_count":64,"outputs":[{"output_type":"display_data","data":{"text/plain":["interactive(children=(IntSlider(value=0, description='instance', max=9), Output()), _dom_classes=('widget-inte…"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"15f701fb400a4f1ab093bdf7a0aceae6"}},"metadata":{}}]}]}