Selma Wanna

Update: I have gotten the following lines of code to run by modifying my LadderNet model to follow this architecture: ``` LadderNetv6( (initial_block): Initial_LadderBlock( (inconv): Conv2d(1, 10, kernel_size=(3, 3), stride=(1,...

Oh, I see. So this framework is not suited for multiclass labels for a single image? I'll review the items you have linked. Thank you very much! I am still...

Hi All, I'm running on Ubuntu 16.04 kernel: 4.13.0-43-generic, and I'm persistently (but at unpredictable times) getting the mismatch error message. I'm not sure if this has been your experience...

Leaving a comment to add that I'm having the same issue.

Any updates on providing the requirements-notorch.txt file?

Hello! I'm in a similar boat. I tried running your script, but for the llama-3-8b model and am having issues as well. I run the following (without internet connection): ```python...

It's possible that I have the same issue. In order to run your code on my system, I have to comment out lines [341](https://github.com/McGill-NLP/llm2vec/blob/72ac29d686a8a5edb7ffa0d7985251cd77bd29bf/llm2vec/llm2vec.py#L341)-361 in llm2vec.py. If I don't comment...

Sure thing! Here is `other_test.py` ```python from llm2vec import LLM2Vec import torch from transformers import AutoTokenizer, AutoModel, AutoConfig from peft import PeftModel from mteb import MTEB MODEL_NAME = "McGill-NLP/LLM2Vec-Meta-Llama-3-8B-Instruct-mntp" #...

Hi all, I have moved my code to an internet-connected, server with 8xH100's. I'm having similar issues with your multigpu .encode() function. See below. I'm still investigating this and don't...