CNN basic

    batch_size = 21
    epochs = 32
    #Complete stock code
    model1 = Sequential()
    model1.add(Conv2D(32, kernel_size=(3,3),activation='relu',input_shape=(35,18,2),padding='valid'))
    model1.add(LeakyReLU(alpha=0.1))
    model1.add(MaxPooling2D(2,padding='same'))
    model1.add(Dropout(0.25))
    model1.add(Conv2D(64, (3,3), activation='relu',padding='same'))
    model1.add(LeakyReLU(alpha=0.1))
    model1.add(MaxPooling2D(pool_size=2,padding='same'))
    model1.add(Dropout(0.25))
    model1.add(Conv2D(128, (3,3), activation='relu',padding='same'))
    model1.add(LeakyReLU(alpha=0.1))                  
    model1.add(MaxPooling2D(pool_size=2,padding='same'))
    model1.add(Dropout(0.4))
    model1.add(Flatten())
    model1.add(Dense(128, activation='relu'))
    model1.add(LeakyReLU(alpha=0.1))       
    model1.add(Dropout(0.3))
    model1.add(Dense(num_classes, activation='softmax'))
    model1.compile(loss=keras.losses.categorical_crossentropy, optimizer=keras.optimizers.Adam(),metrics=['accuracy'])
    model1_train_dropout = model1.fit(train_X, train_Y_one_hot, batch_size=batch_size,epochs=epochs,verbose=1,validation_split=0.2)
    model1.save("sanjoukin_gt_train.h5py")
    #loadedmodel = .load("sanjoukin_gt_train.h5py")
    test_eval = model1.evaluate(test_X, test_Y_one_hot, verbose=11)