-
Hi, sorry to bother you.
I’ve been trying to compile this mod to try it out, but since I’m not familiar with C++ programming, I’m having a lot of trouble compiling the loader (MultiDiva-Loader.dll)…
-
I have meet an error when using demo.py
`(fcclip) ga@test-4U-GPU-Server:~/code/fc-clip$ python demo/demo.py --input 000741.jpg 000860.jpg --opts MODEL.WEIGHTS fcclip_cocopan.pth
[07/15 15:09:43 …
-
### Is your feature request related to a problem? Please describe.
**Editing mode**
![스크린샷 2024-08-23 오후 10 09 17](https://github.com/user-attachments/assets/f05771e8-22f3-492c-af9d-8b5a59…
-
Starting from the tutorial [link](https://github.com/salesforce/ALBEF/blob/main/visualization.ipynb) and considering the function **compute_gradcam** in BlipITM [link](https://github.com/salesforce/LA…
-
- https://arxiv.org/abs/2107.07013
- 2021
過去10年間の機械学習による解釈技術の発展により、人工ニューラルネットワーク(ANN)における分類や定位に最も有益な画像領域を観察する新しいツールが提供されました。
では、同じ領域は、人間の観察者にとっても同様に情報量が多いのでしょうか?
78の実験と6,610人の参加者から得られたデータを用いて、我々…
e4exp updated
3 years ago
-
Hello, I replaced FeatureExtracter in the SLT stage. After training for some time, its accuracy has not improved. I used the best checkpoint to do the test, and every prediction result was consistent.…
-
### Idea Contribution
- [X] I have read all the feature request issues.
- [X] I'm interested in working on this issue
- [X] I'm part of GSSOC organization
### Explain feature request
https://gith…
-
Thanks for this interesting work!
I have a question regarding the feature space alignment. The paper uses the feature in ResNet50 before attention pooling layer, but this feature lives in a different…
-
### Can we access your project?
- [X] I give permission for members of the FlutterFlow team to access and test my project for the sole purpose of investigating this issue.
### Current Behavior
…
-
in the training process, I found that if I set batch size > 1, the loss sometimes will be nan
```
tensor(nan, device='cuda:0', grad_fn=)
```
and some logits also nan
```
(Pdb) p llm_out.logits…