mirrors/mmengine
1
0
Fork 0
mirror of https://github.com/open-mmlab/mmengine.git synced 2025-06-03 21:54:44 +08:00
Code Issues Projects Releases Wiki Activity

[Docs] Fix typo (#1385)

Browse Source
This commit is contained in:
fanqiNO1 2023-10-10 11:32:46 +08:00 committed by GitHub
parent 8015d62202
commit 3b639da1ef
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 4 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
docs/en/common_usage/large_model_training.md
View File

@ -1,4 +1,4 @@
# Traning Big Models
# Training Big Models
When training large models, significant resources are required. A single GPU memory is often insufficient to meet the training needs. As a result, techniques for training large models have been developed, and one typical approach is [DeepSpeed ZeRO](https://www.deepspeed.ai/tutorials/zero/#zero-overview). DeepSpeed ZeRO supports optimizer, gradient, and parameter sharding.
@ -85,7 +85,7 @@ torchrun --nproc-per-node 2 examples/distributed_training_with_flexible_runner.p
```
<details>
<summary>traning log</summary>
<summary>training log</summary>
```
07/03 13:04:17 - mmengine - INFO - Epoch(train) [1][ 10/196] lr: 3.3333e-04 eta: 0:13:14 time: 0.4073 data_time: 0.0335 memory: 970 loss: 6.1887
@ -157,7 +157,7 @@ torchrun --nproc-per-node 2 examples/distributed_training_with_flexible_runner.p
```
<details>
<summary>traning log</summary>
<summary>training log</summary>
```
07/03 13:05:37 - mmengine - INFO - Epoch(train) [1][ 10/196] lr: 3.3333e-04 eta: 0:08:28 time: 0.2606 data_time: 0.0330 memory: 954 loss: 6.1265

2
docs/en/common_usage/save_gpu_memory.md
View File

@ -100,7 +100,7 @@ runner.train()
## Large Model Training
```{warning}
If you have the requirement to train large models, we recommend reading [Traning Big Models](./large_model_training.md).
If you have the requirement to train large models, we recommend reading [Training Big Models](./large_model_training.md).
```
`FSDP` is officially supported from PyTorch 1.11. The config can be written in this way: