CODEWITHSUNDEEP
pythonmachine-learningkeras
Pythoner
Pythoner

Reputation: 127

The results after each training keras model are different

I'm a newbie in Machine Learning. I want to build a keras model which will be used for facial recognition. I am currently using the model at: 

model = Sequential()
model.add(Conv2D(32, (3, 3), padding='same',
             input_shape=x_train.shape[1:]))
model.add(Activation('relu'))
model.add(Conv2D(32, (3, 3)))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))

model.add(Conv2D(64, (3, 3), padding='same'))
model.add(Activation('relu'))
model.add(Conv2D(64, (3, 3)))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))

model.add(Flatten())
model.add(Dense(512))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(num_classes))
model.add(Activation('softmax'))

# initiate RMSprop optimizer
opt = keras.optimizers.rmsprop(lr=0.0001, decay=1e-6)

# Let's train the model using RMSprop
model.compile(loss='categorical_crossentropy',
          optimizer=opt,
           metrics=['accuracy'])

I trained with the same data and parameters the same, but the training results are very different.There are 100% results or 28% results. What made that difference?

Upvotes: 2

Views: 932

Answers (1)

Venkatachalam
Venkatachalam

Reputation: 16966

Setting the seed, when training the model will solve the problem. This will give you the repeatability. 

np.random.seed(10)
tf.set_random_seed(10)

Also make sure train and test split also does not change ever instance. Hence, you can set the seed for data splitting also.

Upvotes: 3

Related Questions