Chengyue Jiang

It will be perfect to have a wrapper function to turn the model into the vllm-enhanced model. (like PEFT). It is useful if we have a lora model, we can...

### Environment GPUs: 4x80G Package Version ------------------------ --------------------- absl-py 2.1.0 aiohttp 3.9.3 aiosignal 1.3.1 appdirs 1.4.4 asttokens 2.4.1 async-timeout 4.0.3 attrs 23.2.0 cachetools 5.3.2 certifi 2024.2.2 charset-normalizer 3.3.2 chex 0.1.82...

I ran the training scripts (using DiT-S/8 by default) successfully, and the loss curve is shown below.  However, the sampled results (also using the default sampling parameters with DiT-XL/2...

Could you provide the dev set?

1

How do you do validation?

[L96](https://github.com/facebookresearch/BLINK/blob/5fe254dd64d37332347edc73738edcb56096183f/blink/biencoder/nn_prediction.py#L96) in [blink/biencoder/nn_prediction.py](https://github.com/facebookresearch/BLINK/blob/main/blink/biencoder/nn_prediction.py) should be changed from ```cur_candidates = candidate_pool[src][inds]``` TO ```cur_candidates = candidate_pool[srcs[i].item()][inds]```

Thanks for your excellent work. There is a typo in the guidelines in the https://github.com/CVI-SZU/Linly/wiki/Huggingface%E6%A0%BC%E5%BC%8F%E8%BD%AC%E6%8D%A2 the --model_name should be changed to --tp_model_name