def get_model(name): #INPUT name: name of the wanted model #OUPUT model: wanted model if name == "tree": model = ExtraTreesRegressor(n_estimators=50) elif name == "line": model = LinearRegression() elif name == "NN": model = Sequential() model.add(Dense(200, input_dim=3)) model.add(Dense(1)) model.compile(optimizer="SGD", loss='mean_squared_error') return model
model.add( Dense(128, kernel_initializer='normal', input_dim=X_train.shape[1], activation='relu')) ##Hidden Layer: model.add(Dense(256, kernel_initializer='normal', activation='relu')) model.add(Dense(256, kernel_initializer='normal', activation='relu')) model.add(Dense(256, kernel_initializer='normal', activation='relu')) #Output layer model.add(Dense(1, kernel_initializer='normal', activation='linear')) #Compile the model model.compile(loss='mean_absolute_error', optimizer='adam', metrics=['mean_absolute_error']) print(model.summary()) ##Now train the model using fit method result = model.fit(X_train, y_train, validation_split=0.3, batch_size=10, epochs=100) #model evaluation prediction = model.predict(X_test) sns.distplot(y_test.values.reshape(-1, 1) - prediction)
Dense(6, input_dim=15, kernel_initializer='random_uniform', activation='relu')) model.add(Dropout(0.2)) model.add( Dense(4, kernel_initializer='random_uniform', activation='relu', kernel_constraint=maxnorm(3))) model.add(Dropout(0.2)) model.add(Dense(2, kernel_initializer='random_uniform', activation='relu')) model.add(Dense(1, kernel_initializer='random_uniform', activation='relu')) # Compile model model.compile(loss='mean_absolute_error', optimizer='adam') # Fit the model model.fit(X_Train, Y_Train, epochs=100, batch_size=10) # Evaluate the model scores = model.evaluate(X_Test, Y_Test) print("score: %.2f%%" % (100 - scores)) model_GT_C_D = 'finalized_model_GT_C_D.sav' pickle.dump(model, open(model_GT_C_D, 'wb')) # Split Data to Train and Test X_Train, X_Test, Y_Train, Y_Test = train_test_split(X, Y2, test_size=0.3) # create model