mirrors
/
fast-reid
mirror of https://github.com/JDAI-CV/fast-reid.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

update readme

pull/153/head
liaoxingyu 2020-07-06 17:04:21 +08:00
parent fec7abc461
commit 13bb8d5b1c
2 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
GETTING_STARTED.md
View File

@ -21,7 +21,7 @@ You may want to use it as a reference to write your own training script.
To train a model with "train_net.py", first setup up the corresponding datasets following [datasets/README.md](https://github.com/JDAI-CV/fast-reid/tree/master/datasets), then run:
```bash
CUDA_VISIBLE_DEVICES=$gpus ./tools/train_net.py --config-file ./configs/Market1501/bagtricks_R50.yml
./tools/train_net.py --config-file ./configs/Market1501/bagtricks_R50.yml MODEL.DEVICE "cuda:0"
```
The configs are made for 1-GPU training.
@ -29,8 +29,8 @@ The configs are made for 1-GPU training.
To evaluate a model's performance, use
```bash
CUDA_VISIBLE_DEVICES=$gpus ./tools/train_net.py --config-file ./configs/Market1501/bagtricks_R50.yml \
--eval-only MODEL.WEIGHTS /path/to/checkpoint_file
./tools/train_net.py --config-file ./configs/Market1501/bagtricks_R50.yml --eval-only \
MODEL.WEIGHTS /path/to/checkpoint_file MODEL.DEVICE "cuda:0"
```
For more options, see `./tools/train_net.py -h`.

4
README.md
View File

@ -4,10 +4,10 @@ FastReID is a research platform that implements state-of-the-art re-identificati
## What's New
- Remove [ignite](https://github.com/pytorch/ignite)(a high-level library) dependency and powered by [PyTorch](https://pytorch.org/).
- [Jul 2020] Distributed training with multiple GPUs, it trains much faster.
- Includes more features such as circle loss, abundant visualization methods and evaluation metrics, SoTA results on conventional, cross-domain, partial and vehicle re-id, testing on multi-datasets simultaneously, etc.
- Can be used as a library to support [different projects](https://github.com/JDAI-CV/fast-reid/tree/master/projects) on top of it. We'll open source more research projects in this way.
- It trains much faster.
- Remove [ignite](https://github.com/pytorch/ignite)(a high-level library) dependency and powered by [PyTorch](https://pytorch.org/).
We write a [chinese blog](https://l1aoxingyu.github.io/blogpages/reid/2020/05/29/fastreid.html) about this toolbox.