-
Just want to say thanks for this! I've been trying to use other peoples' code that all use the `transformer_lens` library and it has a bug that stops you loading models in 4bit, seems to have loads of…
-
## 🐛 Bug
I get a TypeError when trying to load a fairseq model from hub. The same code with `pytorch/vision` and 'resnet50' works. It fails the same with various fairseq models.
## To Reproduce
…
-
As it is right now, when doing finite differences, it's clear that the derivative converges as we go to smaller step sizes. However, when using an odd number of pixels (which at the moment is the corr…
-
## 🐛 Bug
`malloc(): invalid size (unsorted)`
happens with `cudatoolkit==11.0`, works fine with `cudatoolkit==10.2` w/ pt-1.6 or pt-nightly - no difference.
Installed with:
```
conda instal…
-
I was trying to train a VisionEncoderDecoderModel and I got the below error. For decoder I'm using bert-base-multilingual-cased and encoder is google/vit-base-patch16-224. How to solve this error? Tha…
-
While playing around with this library I noticed that the getCurrentArtistAndSong example code would randomly fail occasionally. I turned on debug and noticed that the response codes were still 200s b…
-
### System Info
- `transformers` version: 4.45.1
- Platform: Linux-5.4.0-153-generic-x86_64-with-glibc2.35
- Python version: 3.10.12
- Huggingface_hub version: 0.25.1
- Safetensors version: 0.4.5…
-
-
## 🚀 [RFC] A high-level GSPMD API in PT/XLA (based on `xs.mark_sharding`)
This RFC proposes a high-level API for GSPMD through a wrapper class and a partitioning rule function, based on `xs.mark_sh…
-
Hello! Would it be possible for you to add Mac OS Tome File support to deark? It is at least partially documented in this project:
https://github.com/kainjow/TomeViewerX
Examples are in
htt…