Qi Yan

update crossloc paper (CVPR'22)

https://github.com/zlthinker/KFNet/blob/922ba6430067768da1a14ebd2be44bdecaf43a01/OFlowNet/train.py#L244-L245 The last `_` placeholder is unnecessary and it is inconsistent with `run()`: https://github.com/zlthinker/KFNet/blob/922ba6430067768da1a14ebd2be44bdecaf43a01/OFlowNet/train.py#L198-L199

GPU process issue

5

Hi, Thanks for your wonderful project on dockerizing WeChat and make it runnable on Linux. Things are generally good except for this bug. I saw an bug realted on **GPU...

Hi, Thanks for your thorough literature review and your arxiv paper. Your work helped me a lot when I was first approaching this domain. I wondered if you may accept...

'corrupted size vs. prev_size' error message

2

Hi, thanks for your contribution on this apriltag package. But from time to time, I see the `corrupted size vs. prev_size` error message in the program. It is related to...

Hi, Based on the dataest description on [usage.ipynb](https://github.com/andyzoujm/autocast/blob/f7907d45dd3f5e58d70834f2d7770c404ef4baba/usage.ipynb) notebook, we know that > All numerical questions are (log) normalized to `[0,1]` based on max/min/deriv_ratio stats in `question['choices']`. However, the following...

Hi there, Thanks for open-sourcing the code. I was trying to re-create the retrieved news articles following [retrieve_cc_news_bm25+ce.py](https://github.com/andyzoujm/autocast/blob/master/autocast_experiments/data/retrieve_cc_news_bm25%2Bce.py). However, it is rather slow to run and uses a lot of...

Wrong beta_D coefficient for VP during sampling?

1

Hi, There seems to be a bug in the sampler script. https://github.com/NVlabs/edm/blob/b2a26c921c5776cb52f7498248761d60649007a8/generate.py#L88 https://github.com/NVlabs/edm/blob/b2a26c921c5776cb52f7498248761d60649007a8/generate.py#L91 I suppose the beta_D coefficient should be 19.9. Best, Qi

# 🐛 Bug ## Command I encounter the following error during training from time to time: ``` File "/home/qiyan/anaconda3/envs/mtr/lib/python3.8/site-packages/torch/_tensor.py", line 522, in backward torch.autograd.backward( File "/home/qiyan/anaconda3/envs/mtr/lib/python3.8/site-packages/torch/autograd/__init__.py", line 266, in backward...

1. Hyper-parameter auto-selection randomness. According to P11 of your paper, in each evaluation, two coefficients ε and ε_s are first automatically selected and then used in the Langevin dynamics sampling...