Support loading .pt weights

[shripadk]

Feature request

Need support for loading models that only contain .pt weights

Motivation

I quantized Mixtral 8x7b model using HQQ (which produces a qmodel.pt file). But I am unable to load the weights in LoRAX as it expects either a .safetensors or .bin weights.

Your contribution

I haven't studied the source enough to submit a PR but from cursory understanding of the code, changes need to be made in hub.py file, specifically:

lorax/server/lorax_server/utils/sources/hub.py

Lines 68 to 78 in cc2e0a9

	try:
	filenames = weight_hub_files(model_id, revision, extension, api_token)
	except EntryNotFoundError as e:
	if extension != ".safetensors":
	raise e
	# Try to see if there are pytorch weights
	pt_filenames = weight_hub_files(model_id, revision, extension=".bin", api_token=api_token)
	# Change pytorch extension to safetensors extension
	# It is possible that we have safetensors weights locally even though they are not on the
	# hub if we converted weights locally without pushing them
	filenames = [f"{Path(f).stem.lstrip('pytorch_')}.safetensors" for f in pt_filenames]

Though I would also like to be able to load the base model from local rather than remote/from the hub (as explained in this issue: #347)

[magdyksaleh]

I will work on a fix for this alongside #347