{"cells":[{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nimport glob\nfrom random import shuffle\nfrom sklearn.model_selection import train_test_split\nimport matplotlib.pyplot as plt\nimport tensorflow as tf\nimport math\nimport os\nfrom PIL import Image\nfrom tensorflow.keras import regularizers","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Import TrainDataset and TestDataset"},{"metadata":{"trusted":true},"cell_type":"code","source":"train_dataset=pd.read_csv('../input/fndfeatures6/trainDataset5.csv',sep='\\t')\n#train_dataset=train_dataset.drop('Image_path',axis=1)\ntrain_dataset.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"str1=train_dataset.iloc[0]['properties'][1:-1]\nstr_list=str1.split(',')\nprop=list(map(float,str_list))\nprop","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test_Dataset=pd.read_csv('../input/fndtestdataset/TestDataset.csv',sep=',')\ntest_Dataset.head()\ntest_Features=np.empty((len(test_Dataset),21))\nfor i in range(len(test_Dataset)):\n    str1=test_Dataset.iloc[i]['Features'][1:-1]\n    str_list=str1.split(',')\n    test_Features[i,:]=(list(map(float,str_list)))\n    \ntest_Features[0]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"class DataGenerator1(tf.keras.utils.Sequence):\n    def __init__(self,train_data,batch_size=64,shuffle=True):\n        self.train_data=train_data\n        self.batch_size=batch_size\n        self.shuffle=shuffle\n        self.on_epoch_end()\n        \n    def __len__(self):\n        return (int)(np.floor(len(self.train_data)/self.batch_size))\n    \n    def __getitem__(self,index):\n        batch_ids=self.indexes[index*self.batch_size:(index+1)*self.batch_size]\n        X,y=self.__data_generation(batch_ids)\n        return X,y\n        \n    def on_epoch_end(self):\n        self.indexes=np.arange(len(self.train_data))\n        if(self.shuffle):\n            np.random.shuffle(self.indexes)\n            \n    def __data_generation(self,batch_ids):\n        X=np.empty((len(batch_ids),21))\n        y=np.empty((len(batch_ids),1))\n        for idx,batch_id in enumerate(batch_ids):\n            str1=self.train_data.iloc[batch_id]['properties'][1:-1]\n            str_list=str1.split(',')\n            prop=list(map(float,str_list))\n            X[idx,]=prop\n            y[idx,]=self.train_data.iloc[batch_id]['Class']\n        return X,y","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def DNN_Model():\n    inputs=tf.keras.layers.Input((21))\n    X=tf.keras.layers.BatchNormalization()(inputs)\n    X=tf.keras.layers.Dense(512,activation='relu',kernel_regularizer=regularizers.l2(0.001))(X)\n    X=tf.keras.layers.Dense(256,activation='relu',kernel_regularizer=regularizers.l2(0.001))(X)\n    X=tf.keras.layers.Dropout(0.2)(X)\n    X=tf.keras.layers.BatchNormalization()(X)\n    X=tf.keras.layers.Dense(128,activation='relu',kernel_regularizer=regularizers.l2(0.001))(X)\n    X=tf.keras.layers.Dropout(0.2)(X)\n    X=tf.keras.layers.Dense(64,activation='relu',kernel_regularizer=regularizers.l2(0.001))(X)\n    #X=tf.keras.layers.Dropout(0.2)(X)\n    X=tf.keras.layers.Dense(32,activation='relu',kernel_regularizer=regularizers.l2(0.001))(X)\n   # X=tf.keras.layers.Dropout(0.2)(X)\n    X=tf.keras.layers.BatchNormalization()(X)\n    #X=tf.keras.layers.Dense(24,activation='relu')(X)\n    #X=tf.keras.layers.Dropout(0.2)(X)\n    X=tf.keras.layers.Dense(16,activation='relu',kernel_regularizer=regularizers.l2(0.001))(X)\n    #X=tf.keras.layers.Dropout(0.2)(X)\n    X=tf.keras.layers.Dense(8,activation='relu',kernel_regularizer=regularizers.l2(0.001))(X)   \n    #X=tf.keras.layers.Dropout(0.2)(X)\n    X=tf.keras.layers.BatchNormalization()(X)\n    X=tf.keras.layers.Dense(4,activation='relu',kernel_regularizer=regularizers.l2(0.001))(X)\n    #X=tf.keras.layers.Dropout(0.2)(X)\n    X=tf.keras.layers.Dense(2,activation='relu',kernel_regularizer=regularizers.l2(0.001))(X)\n    #X=tf.keras.layers.Dropout(0.2)(X)\n    outputs=tf.keras.layers.Dense(1,activation='sigmoid')(X)\n    \n    return tf.keras.Model(inputs=inputs,outputs=outputs)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test_data=train_dataset.iloc[0:3000]\ntrain_dataset=train_dataset.iloc[3000:]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test_data.head()\nlen(train_dataset)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_data, val_data=train_test_split(train_dataset,test_size=0.2,random_state=42)\n\ntrain_generator=DataGenerator1(train_data)\nval_generator=DataGenerator1(val_data)\n#X,y=train_generator.__getitem__(0)\n\nmodel=DNN_Model()\nmodel.compile(optimizer='SGD',loss=tf.keras.losses.BinaryCrossentropy(),\\\n              metrics=[tf.keras.metrics.BinaryAccuracy()])\n\nhistory=model.fit_generator(generator=train_generator,validation_data=val_generator\\\n                            ,epochs=300,verbose=1)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Visualize Training process"},{"metadata":{"trusted":true},"cell_type":"code","source":"tf.keras.utils.plot_model(model, show_shapes=True)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"epochs=300\nacc = history.history['binary_accuracy']\nval_acc = history.history['val_binary_accuracy']\n\nloss = history.history['loss']\nval_loss = history.history['val_loss']\n\nepochs_range = range(epochs)\n\nplt.figure(figsize=(8, 8))\nplt.subplot(1, 2, 1)\nplt.plot(epochs_range, acc, label='Training Accuracy')\nplt.plot(epochs_range, val_acc, label='Validation Accuracy')\nplt.legend(loc='lower right')\nplt.title('Training and Validation Accuracy')\n\nplt.subplot(1, 2, 2)\nplt.plot(epochs_range, loss, label='Training Loss')\nplt.plot(epochs_range, val_loss, label='Validation Loss')\nplt.legend(loc='upper right')\nplt.title('Training and Validation Loss')\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test_X=[]\ntest_label=[]\nfor i in range(3000):\n    str1=test_data.iloc[i]['properties'][1:-1]\n    str_list=str1.split(',')\n    prop=list(map(float,str_list))\n    test_X.append(prop)\n    test_label.append(test_data.iloc[i]['Class'])\n    \ntest_X=np.array(test_X)\ntest_label=np.array(test_label)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"y_pred=model.predict(test_X)\npred_y=[]\nfor i in range(len(y_pred)):\n    if(y_pred[i]>=0.5):\n        pred_y.append(1)\n    else:\n        pred_y.append(0)\n        \npred_y=np.array(pred_y)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from sklearn import metrics\nmetrics.accuracy_score(test_label,pred_y)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"metrics.f1_score(test_label,pred_y)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Get Predictions of TestDataset"},{"metadata":{"trusted":true},"cell_type":"code","source":"test_y=model.predict(test_Features)\ny_test=[]\nfor i in range(len(test_y)):\n    if(test_y[i]>=0.5):\n        y_test.append(1)\n    else:\n        y_test.append(0)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"submission=pd.DataFrame(\n    {'ImageId':test_Dataset['Iamge_Id'],\n    'label':y_test})\nsubmission.head()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Save model and submission file"},{"metadata":{"trusted":true},"cell_type":"code","source":"submission.to_csv('./submission.csv',index=False)\nmodel.save('./MLP10.h5')","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat":4,"nbformat_minor":1}