{ "metadata": { "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.8.5-final" }, "orig_nbformat": 2, "kernelspec": { "name": "python_defaultSpec_1600382692523", "display_name": "Python 3.8.5 64-bit" } }, "nbformat": 4, "nbformat_minor": 2, "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Trainging a Classifier\n", "After getting familiar with the data, we want to train a model which shall classify the data. Since the data is labeled, we can run a supervised training\n", "\n", "## Loading the Data\n", "As shown before, we need to loda the dataframe first:" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "\n", "data_directory = 'http://hadron.physics.fsu.edu/~dlersch/GlueX_PANDA_EIC_ML_Workshop'\n", "data_name = 'hands_on_data_033_033_033.csv'\n", "dataFrame = pd.read_csv(data_directory + '/' + data_name)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Setting up two Classifier\n", "Now we make use of the fast scikit library which offers many algorithms. A detailed describtion of the classifier available can be found here: https://scikit-learn.org/stable/supervised_learning.html#supervised-learning For now, we will just train two classification algorithms: i) Neural Network and ii) A radnom forest classifier. These are the most promiment examples for solving classification problems. You are of course welcome to test/choose any other classification algorithm. " ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "tags": [] }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": "\nSetting up a multilayer perceptron...\n...done!\n \n \nSetting up a radnom forest...\n...done!\n \n" } ], "source": [ "from sklearn.utils import shuffle\n", "from sklearn.neural_network import MLPClassifier\n", "from sklearn.ensemble import RandomForestClassifier\n", "\n", "#1.) Setting up the neural network:\n", "print(\" \")\n", "print(\"Setting up a multilayer perceptron...\")\n", "\n", "use_mlp_early_stopping = True #Change this flag to train the MLP with / without early stopping\n", "my_mlp = MLPClassifier(\n", " hidden_layer_sizes=(5),\n", " activation='tanh',\n", " solver='sgd',\n", " shuffle=True,\n", " validation_fraction=0.25,#--> Use validation data to avoid overfitting\n", " early_stopping=use_mlp_early_stopping, #--> Change to disable early stopping and see the difference, no validation curve available then...\n", " max_iter = 10,\n", " learning_rate_init=0.01,\n", " warm_start=True,\n", " tol=1e-6\n", ")\n", "print(\"...done!\")\n", "print(\" \")\n", "\n", "#2.) Setting up the random forest:\n", "print(\" \")\n", "print(\"Setting up a radnom forest...\")\n", "my_rf = RandomForestClassifier(\n", " n_estimators=10, #--> Number of trees in your forest\n", " warm_start=True,\n", " max_depth=5, #--> Maximum depth of tree\n", " bootstrap=True, #--> Sample subsets from the training data to train each tre with an indiviudal set\n", " random_state=0\n", ")\n", "print(\"...done!\")\n", "print(\" \")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Preparing the Data for Training\n", "After setting up the classifier, we need to prepare the data properly, i.e. define the input features (X) and targets (Y). This can be easily done using the pandas librabry. Finally, we normalize and shuffle the data. " ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "from sklearn.preprocessing import MinMaxScaler\n", "\n", "used_features = ['var1','var2','var3','var4'] #--> Change the elements here, in order to use different features\n", "X = dataFrame[used_features].values\n", "Y = dataFrame['label'].values\n", "\n", "scaler = MinMaxScaler() \n", "X = scaler.fit_transform(X) #Uncomment this line to run without feature normalization\n", "\n", "x_train, y_train = shuffle(X,Y,random_state=0)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Training the Neural Network\n", "First, we train the multilayer perceptron (MLP) and check the training curve(s) afterwards:" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "tags": [] }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": "\nTrain MLP...\n...done!\n \n" }, { "output_type": "display_data", "data": { "text/plain": "
", "image/svg+xml": "\n\n\n\n \n \n \n \n 2020-09-17T18:47:12.161050\n image/svg+xml\n \n \n Matplotlib v3.3.0, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n", "image/png": "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\n" }, "metadata": { "needs_background": "light" } } ], "source": [ "import matplotlib.pyplot as plt\n", "\n", "print(\" \")\n", "print(\"Train MLP...\")\n", "my_mlp.fit(x_train,y_train)\n", "print(\"...done!\")\n", "print(\" \")\n", "\n", "#Getting the training/validation scores is quite easy:\n", "plt.rcParams.update({'font.size': 25})\n", "\n", "training_curve = my_mlp.loss_curve_\n", "plt.plot(training_curve,label='training data')\n", "\n", "if use_mlp_early_stopping:\n", " validation_curve = my_mlp.validation_scores_\n", " plt.plot(validation_curve,label='validation data')\n", "\n", "plt.legend()\n", "plt.xlabel('Epoch')\n", "plt.ylabel('Loss')\n", "plt.ylim(0.0,1.0)\n", "plt.grid(True)\n", "plt.show()\n", "plt.close()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Finally, we write the mlp model to a file, because we want to use it later many times." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": "['mlp_classifier.joblib']" }, "metadata": {}, "execution_count": 5 } ], "source": [ "from joblib import dump\n", "dump(my_mlp,'mlp_classifier.joblib')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Training the Random Forest Classifier\n", "In a next step we traing the random forest and store it afterwards" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "tags": [] }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": "\nTrain random forest classifier...\n...done!\n \n" }, { "output_type": "execute_result", "data": { "text/plain": "['random_forest_classifier.joblib']" }, "metadata": {}, "execution_count": 6 } ], "source": [ "print(\" \")\n", "print(\"Train random forest classifier...\")\n", "my_rf.fit(x_train,y_train)\n", "print(\"...done!\")\n", "print(\" \")\n", "\n", "dump(my_rf,'random_forest_classifier.joblib')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Checking the Classifier Predictions\n", "As stated earlier, one of the very first steps after training your model is to check the classifier outputs and the features after classification. First we add the classsifier predictions to the dataframe. This is not always necessary, but it might be sometimes helfpul to store the dataframe including the predictions somewhere. " ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "tags": [] }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": "\nAdd classifier predictions to the dataframe...\n...done!\n \nJust to be sure, check the dataframe:\n var1 var2 var3 var4 var5 var6 label \\\n0 1.058485 0.966484 0.018853 0.004739 0.947370 0.718451 2.0 \n1 4.592704 3.808958 0.021037 0.005426 3.360293 0.481386 2.0 \n2 1.361142 0.213187 0.003005 0.003969 0.243650 0.325567 0.0 \n3 2.204165 0.969037 0.007780 0.016829 0.719581 0.734593 1.0 \n4 0.629281 0.618626 0.024197 0.006617 0.637974 0.576872 2.0 \n5 1.048963 0.279840 0.002310 0.003568 0.191373 0.665752 0.0 \n6 0.896067 0.329471 0.005907 0.005206 0.325601 0.705401 0.0 \n7 0.873841 0.489065 0.018118 0.013656 0.364731 0.681083 1.0 \n8 0.330018 0.372559 0.020232 0.022098 0.312272 0.210790 1.0 \n9 1.014085 0.706302 0.014414 0.005087 0.791526 0.434839 2.0 \n\n MLP_Output1 MLP_Output2 MLP_Output3 RF_Output1 RF_Output2 RF_Output3 \n0 3.605843e-07 0.000832 0.999168 0.009734 0.020084 0.970182 \n1 6.523371e-02 0.260941 0.673825 0.030361 0.043406 0.926233 \n2 9.996643e-01 0.000310 0.000026 0.983736 0.009596 0.006667 \n3 5.380765e-03 0.971326 0.023293 0.050121 0.945895 0.003984 \n4 3.493463e-06 0.008625 0.991371 0.009112 0.021291 0.969597 \n5 9.986143e-01 0.000974 0.000412 0.981467 0.011958 0.006574 \n6 9.955549e-01 0.002956 0.001489 0.976572 0.015295 0.008133 \n7 5.279794e-04 0.994403 0.005069 0.007614 0.981677 0.010710 \n8 1.976461e-03 0.997931 0.000093 0.006337 0.984347 0.009316 \n9 1.736347e-03 0.021663 0.976600 0.025351 0.041162 0.933487 \n \n" } ], "source": [ "print(\" \")\n", "print(\"Add classifier predictions to the dataframe...\")\n", "\n", "mlp_predictions = my_mlp.predict_proba(X)\n", "rf_predictions = my_rf.predict_proba(X)\n", "\n", "dataFrame['MLP_Output1'] = mlp_predictions[:,0]\n", "dataFrame['MLP_Output2'] = mlp_predictions[:,1]\n", "dataFrame['MLP_Output3'] = mlp_predictions[:,2]\n", "\n", "dataFrame['RF_Output1'] = rf_predictions[:,0]\n", "dataFrame['RF_Output2'] = rf_predictions[:,1]\n", "dataFrame['RF_Output3'] = rf_predictions[:,2]\n", "\n", "print(\"...done!\")\n", "print(\" \")\n", "\n", "print(\"Just to be sure, check the dataframe:\")\n", "print(dataFrame.head(10))\n", "print(\" \")\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now plot the outpus for each species and each classifier:" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "output_type": "display_data", "data": { "text/plain": "
", "image/svg+xml": "\n\n\n\n \n \n \n \n 2020-09-17T18:48:07.167939\n image/svg+xml\n \n \n Matplotlib v3.3.0, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n", "image/png": "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\n" }, "metadata": { "needs_background": "light" } } ], "source": [ "fig,ax = plt.subplots(1,2)\n", "fig.set_size_inches(15, 8)\n", "\n", "ax[0].hist(dataFrame[dataFrame['label']==0.0]['MLP_Output1'],bins=100,facecolor='g',alpha=0.5,label='Species 1',log=True)\n", "ax[0].hist(dataFrame[dataFrame['label']==1.0]['MLP_Output2'],bins=100,facecolor='r',alpha=0.5,label='Species 2',log=True)\n", "ax[0].hist(dataFrame[dataFrame['label']==2.0]['MLP_Output3'],bins=100,facecolor='b',alpha=0.5,label='Species 3',log=True)\n", "\n", "ax[1].hist(dataFrame[dataFrame['label']==0.0]['RF_Output1'],bins=100,facecolor='g',alpha=0.5,label='Species 1',log=True)\n", "ax[1].hist(dataFrame[dataFrame['label']==1.0]['RF_Output2'],bins=100,facecolor='r',alpha=0.5,label='Species 2',log=True)\n", "ax[1].hist(dataFrame[dataFrame['label']==2.0]['RF_Output3'],bins=100,facecolor='b',alpha=0.5,label='Species 3',log=True)\n", "\n", "ax[0].set_xlabel('Neural Network Outputs')\n", "ax[0].set_ylabel('Entries [a.u.]')\n", "ax[0].legend()\n", "\n", "ax[1].set_xlabel('Random Forest Outputs')\n", "ax[1].legend()\n", "\n", "plt.show()\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now we will look at the feature distributions after classfying them with the MLP. We will not select a certain species (you can do that...), but look at the entire data, after applying a threshold to the mlp output 1" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "output_type": "display_data", "data": { "text/plain": "
", "image/svg+xml": "\n\n\n\n \n \n \n \n 2020-09-17T18:48:17.741673\n image/svg+xml\n \n \n Matplotlib v3.3.0, https://matplotlib.org/\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n", "image/png": "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\n" }, "metadata": { "needs_background": "light" } } ], "source": [ "from matplotlib.colors import LogNorm\n", "\n", "fig_data,ax_data = plt.subplots(1,2)\n", "fig_data.set_size_inches(25, 8)\n", "\n", "threshold = 0.5\n", "\n", "ax_data[0].hist2d(dataFrame[dataFrame['MLP_Output1'] > threshold]['var1'],dataFrame[dataFrame['MLP_Output1'] > threshold]['var2'],bins=100,norm=LogNorm())\n", "ax_data[1].hist2d(dataFrame[dataFrame['MLP_Output1'] > threshold]['var3'],dataFrame[dataFrame['MLP_Output1'] > threshold]['var4'],bins=100,norm=LogNorm())\n", "\n", "ax_data[0].set_xlabel('Variable 1')\n", "ax_data[0].set_ylabel('Variable 2')\n", "\n", "ax_data[1].set_xlabel('Variable 4')\n", "ax_data[1].set_ylabel('Variable 3')\n", "\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "NOTE: You can get the label directly by calling: your_model.predict(X). The predicted label will correspond to the number of the maximum output value. Feel free to explore this function and add it to the dataframe." ] } ] }