Validation split argument

[micah5]

Describe the feature you'd like
An input argument to fit() that automatically splits the training data into train & validation sets.

How would this feature be used? Please describe.

Input float between 0 and 1 to fraction of the training data to be used as validation data.

i.e. instead of

model.fit(inputs={'train': s3_input("s3://.../train"), 'validation': s3_input("s3://.../valid")}) 

it would be something like:

model.fit(inputs={'train': s3_input("s3://.../train"), 'validation_split': 0.2}) 

and 20% of the training data would be held away as a validation set.

Describe alternatives you've considered
See the implementation in keras: https://keras.rstudio.com/reference/fit.html

[akrishna1995]

We have added your feature request to our backlog of feature requests and may consider putting it into future SDK versions. I will go ahead and close the issue now, please let me know if you have any more feedback.