github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-09-22 09:59:38 +00:00
Code Issues Packages Projects Releases Wiki Activity

[doc] update booster tutorials (#3718)

Browse Source 
* [booster] update booster tutorials#3717

* [booster] update booster tutorials#3717, fix

* [booster] update booster tutorials#3717, update setup doc

* [booster] update booster tutorials#3717, update setup doc

* [booster] update booster tutorials#3717, update setup doc

* [booster] update booster tutorials#3717, update setup doc

* [booster] update booster tutorials#3717, update setup doc

* [booster] update booster tutorials#3717, update setup doc

* [booster] update booster tutorials#3717, rename colossalai booster.md

* [booster] update booster tutorials#3717, rename colossalai booster.md

* [booster] update booster tutorials#3717, rename colossalai booster.md

* [booster] update booster tutorials#3717, fix

* [booster] update booster tutorials#3717, fix

* [booster] update tutorials#3717, update booster api doc

* [booster] update tutorials#3717, modify file

* [booster] update tutorials#3717, modify file

* [booster] update tutorials#3717, modify file

* [booster] update tutorials#3717, modify file

* [booster] update tutorials#3717, modify file

* [booster] update tutorials#3717, modify file

* [booster] update tutorials#3717, modify file

* [booster] update tutorials#3717, fix reference link

* [booster] update tutorials#3717, fix reference link

* [booster] update tutorials#3717, fix reference link

* [booster] update tutorials#3717, fix reference link

* [booster] update tutorials#3717, fix reference link

* [booster] update tutorials#3717, fix reference link

* [booster] update tutorials#3717, fix reference link

* [booster] update tutorials#3713

* [booster] update tutorials#3713, modify file
This commit is contained in:
jiangmingyan
2023-05-18 11:41:56 +08:00
committed by GitHub
parent 05759839bd
commit d449525acf
7 changed files with 213 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
docs/source/zh-Hans/basics/launch_colossalai.md
View File

@@ -74,7 +74,7 @@ import colossalai
args = colossalai.get_default_parser().parse_args()
# launch distributed environment
colossalai.launch(config=<CONFIG>,
colossalai.launch(config=args.config,
rank=args.rank,
world_size=args.world_size,
host=args.host,
@@ -93,12 +93,21 @@ PyTorch自带的启动器需要在每个节点上都启动命令才能启动多
首先，我们需要在代码里指定我们的启动方式。由于这个启动器是PyTorch启动器的封装，那么我们自然而然应该使用`colossalai.launch_from_torch`。
分布式环境所需的参数，如 rank, world size, host 和 port 都是由 PyTorch 启动器设置的，可以直接从环境变量中读取。
config.py
```python
BATCH_SIZE = 512
LEARNING_RATE = 3e-3
WEIGHT_DECAY = 0.3
NUM_EPOCHS = 2
```
train.py
```python
import colossalai
colossalai.launch_from_torch(
config=<CONFIG>,
config="./config.py",
)
...
```
接下来，我们可以轻松地在终端使用`colossalai run`来启动训练。下面的命令可以在当前机器上启动一个4卡的训练任务。