-
When using float32 as computing datatype, after training steps are complete I get the error message:
ValueError: You cannot perform fine-tuning on purely quantized models. Please attach trainable a…
-
hello, in **quant_train_module.py** file, i saw a line of code : y.data.copy_(yq.data), this code change the data of relu's output data.data, in order to use it in backword for calculate activation's…
-
With support for [metadata in 3D Tiles Next](https://github.com/CesiumGS/3d-tiles/tree/main/next#metadata) comes new options for storing large, columnar property arrays in binary form. Metadata descri…
-
"self.current_input_max = F.max(F.abs(x), axis=(1, 2, 3)).mean().asscalar()" inside "def _conv2d_forward", this will create symbol node, and should cause error because asscalar() is not supported in s…
-
### System Info
```Shell
`Accelerate` version: 0.31.0
- Platform: Linux-5.4.0-171-generic-x86_64-with-glibc2.35
- `accelerate` bash location: /workspace/Thesis/venv/bin/accelerate
- Python vers…
-
**Proceedings**
https://papers.nips.cc/book/advances-in-neural-information-processing-systems-30-2017
https://github.com/catpanda/NIPS_2017
**PaperLists (#Papers 679)**
https://www.dropbox.com/s…
-
-
With the default parameters (simply cloning the repo and training), the model could not fit a sine wave (I was intentionally trying to overfit on one of the simplest examples as a sanity check).
The …
-
- [ ] [self-speculative-decoding/README.md at main · dilab-zju/self-speculative-decoding](https://github.com/dilab-zju/self-speculative-decoding/blob/main/README.md?plain=1)
# Self-Speculative Decod…
-
- [ ] [Answer.AI - You can now train a 70b language model at home](https://www.answer.ai/posts/2024-03-06-fsdp-qlora.html)
# Answer.AI - You can now train a 70b language model at home
**DESCRIPTION:…