-
Hello, I am trying to recreate this notebook https://colab.research.google.com/github/huggingface/blog/blob/master/notebooks/01_how_to_train.ipynb for transformer XL
I made changes to the tokenizer a…
-
微调脚本 **finetune/34B/finetune.sh** 中:
--learning_rate 9.65e-6 \
--lr_scheduler_type 'linear' \
**请问**:以上两个参数是经验总结的最优值吗?或者是针对34B参数的最优值吗?
很多微调脚本中(比如FastChat项目) lr_scheduler_type 设置的都是 cosine,learni…
-
## Bug Description
I am receving this error when compling a torchscript module. The module is a light wrapper around a [SequenceGenerator](https://github.com/facebookresearch/fairseq/blob/main/fair…
-
Hi,
Its gonna be hard for me to pinpoint this broken inference because I am not getting any logs except sometimes I get CUDA errors. But its intermittent. After the new version release I have have …
-
### System Info
- `transformers` version: 4.31.0
- Platform: Linux-5.19.0-42-generic-x86_64-with-glibc2.35
- Python version: 3.9.16
- Huggingface_hub version: 0.14.1
- Safetensors version: 0.3.1
…
-
请问出现这个问题是什么原因呢?
```
2023-02-09 08:23:10 | INFO | fairseq.tasks.translation | /data/home/likai/NMT-offline/knn-box/knnbox-scripts/vanilla-knn-mt/../../data-bin/zh2en-ziyan-03 train zh-en 721 exampl…
-
Please find the (inofficial) environment.yml below. I had some issues setting up the environment correctly myself, and hope this will help someone else.
DISCLAIMER: I have not yet checked if the m…
-
My minimal example:
```python
import torch
device = "cuda" if torch.cuda.is_available() else "cpu"
from transformers import AutoTokenizer, StoppingCriteria, StoppingCriteriaList
repo = "meta-llam…
-
Does the model can be converted to onnx model?
-
我修改了streamllama.py文件chat_completion,执行完for循环后直接返回结果,然后直接return结果,但是这样第一次调用是没问题,第二次之后就会超时失败了