Roy McClanahan

Results 5 comments of Roy McClanahan

How to use endpoint for model published on EC-2 server

Also, see this generic interface I found very useful for something like this: https://dspy-docs.vercel.app/docs/deep-dive/language_model_clients/custom-lm-client

bart-large-xsum model: There were missing keys in the checkpoint model loaded: ['model.encoder.embed_tokens.weight', 'model.decoder.embed_tokens.weight', 'lm_head.weight'].

I have run into this exact same issue today (with the exact same list of missing fields) but with a different model `facebook/nllb-200-distilled-600M`. I took the time to create a...

bart-large-xsum model: There were missing keys in the checkpoint model loaded: ['model.encoder.embed_tokens.weight', 'model.decoder.embed_tokens.weight', 'lm_head.weight'].

@ArthurZucker I upgraded per your recommendation to the latest versions of safetensors 0.4.2 and transformers 4.39.1 and the issue continues to happen. I will try to find time to boil...

bart-large-xsum model: There were missing keys in the checkpoint model loaded: ['model.encoder.embed_tokens.weight', 'model.decoder.embed_tokens.weight', 'lm_head.weight'].

@ArthurZucker I noticed something just now that probably explains why this is a rarer behavior to see. I am using `load_best_model_at_end=True` for my `Seq2SeqTrainingArguments`. When I switch this to `load_best_model_at_end=False`,...

Improve Next.js support

Since writing a set of wrappers for all MUI components that include the `use client` directive appears to be the solution most people are going with, has anyone produced such...