Factor out backend (eg PyTorch, TensorFlow) validation.

[arnaudvl]

Add _validate_backend functionality instead of checking on the fly as in https://github.com/SeldonIO/alibi-detect/blob/master/alibi_detect/cd/classifier.py#L98.

[amanjha8100]

I would like to solve this issue. I am new to open source therefore If I make any mistakes please do guide me.

[jklaise]

@amanjha8100 for sure! The most appropriate place for this function seems to be https://github.com/SeldonIO/alibi-detect/blob/master/alibi_detect/utils/frameworks.py. The function should take backend string provided by the user and validate it pretty much the same way as these lines:

alibi-detect/alibi_detect/cd/classifier.py

Lines 97 to 102 in 13b2664

	backend = backend.lower()
	if backend == 'tensorflow' and not has_tensorflow or backend == 'pytorch' and not has_pytorch:
	raise ImportError(f'{backend} not installed. Cannot initialize and run the '
	f'ClassifierDrift detector with {backend} backend.')
	elif backend not in ['tensorflow', 'pytorch']:
	raise NotImplementedError(f'{backend} not implemented. Use tensorflow or pytorch instead.')

It would also be useful to define a BACKENDS=['tensorflow', 'pytorch'] constant in this module.

@arnaudvl we may need additional functionality in the future, e.g. if some methods implement one backend but not the other.

[amanjha8100]

I will try to solve this issue. Thank you for the guidance.

[ascillitoe]

This issue will be solved by #537.