NbAiLabArchive
/

test_w8

Model card Files Files and versions

Metrics Training metrics Community

test_w8 / generate_pytorch_model.py

pere's picture

Saving weights and logs of step 5000

095f5a3 over 4 years ago

history blame contribute delete

256 Bytes

	# This script overwrites any existing PyTorch model. Generates a new one with an LM head from the pretrained Flax model.
	from transformers import RobertaForMaskedLM
	model = RobertaForMaskedLM.from_pretrained(".",from_flax=True)
	model.save_pretrained(".")