CODEWITHSUNDEEP
pythontensorflowmachine-learningkerasdeep-learning
Nadeem Yousaf
Nadeem Yousaf

Reputation: 11

Can a Model maintain the order of prediction in which the features were given to it

Can a model maintain the order of predictions in loss function based on the order of the batch of input features given to it while training.? if not how can i make it remember the order because i need this order in my custom loss function currently its failing which i think is because its not following the order. for example input features are 45,12,23 and ground truth of 1,2,3 so in predictions i want the same order Such that loss function gets the predictions in the same order as the input features were.

Upvotes: 1

Views: 745

Answers (1)

Shubham Panchal
Shubham Panchal

Reputation: 4289

If you're using tf.keras.Sequential, the fit() method has a shuffle argument which is set to True by default.

You may set to False to avoid shuffling of samples. Note, this argument has no effect when steps_per_epoch argument is None. See the docs.

Upvotes: 1

Related Questions