-
This is my scripts:
```
torchrun --nnodes 1 \
--nproc_per_node 8 \
-m open_clip_train.main \
--model RN50 \
--train-data 'datasets/cc3m/cc3m-train-{0000..0575}.tar' \
--trai…
-
Cool paper and model! I would really like to do some benchmarking of my own, but it seems like you have some missing files. E.g. in notebook 6_:
https://github.com/a96123155/UTR-LM/blob/main/Scripts/…
-
Hi,
I try to load the pre-trained models for inference:
CT_CLIP_zeroshot.pt,
CT_LiPro.pt
I run:
run_zero_shot.py with these parameters:
clip.load(r"C:\Users\pretrained_models\CT_CLIP_zeroshot…
-
I want to ask that for a few datasets, I followed https://github.com/openai/CLIP/blob/main/notebooks/Prompt_Engineering_for_ImageNet.ipynb
to reproduce the results in table 11. I could produce the re…
-
In UniTS_zeroshot_newdata.sh，we found that we need to pre-train before we could test zero-shot. Is there a pre-trained model on this for direct testing?
-
I know that the SRL tags are different in the current version of the SRL software employed. However, I noticed that besides this detail, it seems that there is another discrepancy between versions. Th…
-
---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
Cell In[22], line 1
----> 1 zeroshot_trainer.…
-
# Seoyeon’s Blog for study - [LLM]Zeroshot
[https://seoyeonc.github.io/blog/posts/RESEARCHES/2023-08-26-zero-shot.html](https://seoyeonc.github.io/blog/posts/RESEARCHES/2023-08-26-zero-shot.html)
-
I was advised that given the long strings of reasoning data I am using to SFT my models, that I should run the MMLU benchmark with CoT. Any advise how to do this?
I put the few-shot flag but it di…
-
when executing zero shot. the following probs is empty:
topics, probs = topic_model.fit_transform(docs, embeddings)
following configuration:
from umap import UMAP
umap_model = UMAP(n_neighbo…