Find answers from the community

Updated 11 months ago

Optimum

@Logan M Trying to use Optimum ONNX Embedding for bge as shown in the documentation examples on my Macbook with M1 Pro. I get this error when I try to test the model with get_text_embedding : InvalidArgument: [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Invalid Feed Input Name:token_type_ids. No clue how to debug and fix this, tried looking it up online to understand little to nothing about this

8 comments

LLogan M

Classic huggingface. Either they updated the library or the model I picked with the notebook I just got lucky.

It's raising an error because token_type_ids is being passed in, but it's unused

Probably need to remove it from the tokenizer output here

https://github.com/run-llama/llama_index/blob/92f82f83f5dc4ea9f236eff066e53df264a8c1f1/llama_index/embeddings/huggingface_optimum.py#L133

VVish

Ah, that makes sense lol

VVish

On an unrelated note @Logan M would I be able to use OpenRouter to access Mistral 8x7B via the OpenAI Like LLM?

LLogan M

Maybe!

That reminds me, I need to finish the PR, they guy stopped replying lol

https://github.com/run-llama/llama_index/pull/9464

VVish

Looking forward to the finished PR!

VVish

Lmk if you think I can do anything to help

LLogan M

Merged open router!

VVish

Thank you so much!

Add a reply