Closed dinhngoc267 closed 4 months ago
Hi, after pre-training t5 by nanot5, I want to use it's weight into my custom finetunning code for my task.
When I load the pretrained weight,
t5 = T5ForConditionalGeneration.from_pretrained('nanoT5/nanoT5/logs/2024-07-09/02-39-25-/checkpoint-pt-151', return_dict=True)
it raised me an error:
checkpoint-pt-151 does not appear to have a file named config.json
How to solve this. Thank you
Oh I have to download the base model config and put in the checkpoint folder and it works!
Hi, after pre-training t5 by nanot5, I want to use it's weight into my custom finetunning code for my task.
When I load the pretrained weight,
it raised me an error:
How to solve this. Thank you