-
```
nvcc -v
```
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2022 NVIDIA Corporation
Built on Wed_Sep_21_10:41:10_Pacific_Daylight_Time_2022
Cuda compilation tools, release 11.8, V11…
-
编译时遇到问题:
TypeError: expected string or bytes-like object
[end of output]
note: This error originates from a subprocess, and is likely not a problem with pip.
error: legacy-install-failure
**OS:…
-
### Project URL
https://test.pypi.org/project/bmtrain/0.2.2/
https://pypi.org/project/bmtrain/0.2.2/
### Does this project already exist?
- [X] Yes
### New Limit
500MB
### Update is…
-
When I run the following command to install BMTrain
`python setup.py install`
I met the error of
```
running install
/home/yhlin/torch_env/lib/python3.8/site-packages/setuptools/command/instal…
-
具体报错信息:
```shell
>>> import bmtrain as bmt
Traceback (most recent call last):
File "", line 1, in
File "/FlagAI/examples/Aquila/Aquila-chat/BMTrain/bmtrain/__init__.py", line 2, in
fro…
-
I have tried to load the checkpoints through:
```sh
local_rank = int(os.environ["LOCAL_RANK"])
bmtrain.init_distributed(seed=seed)
rank_print('loading tokenizer...')
tokenizer…
-
We want to try a large LM model (>30B).
Are there any examples to do that?
-
torch 2.0.1+cu117
ubuntu22
安装bmtrain成功,import报错:
Traceback (most recent call last):
File "", line 1, in
File "/home/kidd/anaconda3/envs/cpmbee/lib/python3.8/site-packages/bmtrain/__init__.py…
-
CUDA环境变量也配置没问题,但还是提示这个,我把cuda版本从12.1降到11.8都是同样的错误
python setup.py install
No CUDA runtime is found, using CUDA_HOME='/usr/local/cuda'
running install
/root/anaconda3/envs/cpm-bee/lib/python3.8/sit…
-
In the environment of pytorch 1.12.1, cuda 11.1, and python 3.8.0, I failed to install BMTrain 0.2.0 using "pip setup.py install". Prompt that cuda needs to meet version 11.3. Does BMTrain 0.2.0 suppo…