realgump
realgump
### Checklist - I have searched related issues but cannot get the expected help. - I have read related documents and don't know what to do. ### Describe the question...
I try to load _vivit_base_fe_ model using _flax.training_, and find the numbers of layers of SpatialTransformer and TemporalTransformer are **both 12.** However, when I check [vivit_base_factorised_encoder](https://github.com/google-research/scenic/blob/7d1a639c969a7ba03d70af4ee571e65084fe1a2b/scenic/projects/vivit/configs/kinetics400/vivit_base_factorised_encoder.py), I find **config.model.temporal_transformer.num_layers =...
Hello, thanks for your great work. I have successfully trained the Vivit. However, only several checkpoints are available. In another [issue](https://github.com/mx-mark/VideoTransformer-pytorch/issues/19#issuecomment-1133498281), you have mentioned that the pre-trained models are from...
_self.alpha_ seems no need to update when forward, which raises "out of index" error when **batch_size < class_num**.
### Issue you'd like to raise. When I using agent with a llm of gpt-3.5 and a search tool of google, the AI's response is always in English, regardless of...
I tried to infer my data using `get_model_answer.py` with A100-80g, but each query took over 30 seconds to infererence. However, when I deployed the model with openai-api on the same...
您好，论文表里面的zero-1和dp的激活值为什么是A/N呢？如果是这样的话，sp+zero-1相比zero-1似乎没有优势。