ayo/ipex-llm
1
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

Update pytorch to 1.11 (#4854)

Browse source 
* update setup and action

* fix yml

* update yml

* test yml

* update docs

* update notebook requirements

* test notebooks unit test

* reset yml

* delete comments

* fix yml

* fix errors in setup.py

* fix setup.py

* specify the version of IPEX
This commit is contained in:
Mingzhi Hu 2022-06-24 16:38:48 +08:00 committed by GitHub
parent 8e10bdeabd
commit 145216bfc1
1 changed files with 0 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

7
docs/readthedocs/source/doc/Nano/QuickStart/pytorch_train.md
View file

@ -47,13 +47,6 @@ from bigdl.nano.pytorch import Trainer
trainer = Trainer(max_epoch=10, use_ipex=True) trainer = Trainer(max_epoch=10, use_ipex=True)
``` ```
Note: BigDL-Nano does not install IPEX by default. You can install IPEX using the following command:
```bash
python -m pip install torch_ipex==1.9.0 -f https://software.intel.com/ipex-whl-stable
python -m pip install torchvision==0.10.0+cpu -f https://download.pytorch.org/whl/torch_stable.html
```
#### Multi-instance Training #### Multi-instance Training
When training on a server with dozens of CPU cores, it is often beneficial to use multiple training instances in a data-parallel fashion to make full use of the CPU cores. However, using PyTorch's DDP API is a little cumbersome and error-prone, and if not configured correctly, it will make the training even slow. When training on a server with dozens of CPU cores, it is often beneficial to use multiple training instances in a data-parallel fashion to make full use of the CPU cores. However, using PyTorch's DDP API is a little cumbersome and error-prone, and if not configured correctly, it will make the training even slow.