Skip to content

Latest commit

 

History

History
42 lines (27 loc) · 1.47 KB

GETTING_STARTED.md

File metadata and controls

42 lines (27 loc) · 1.47 KB
Raw

Getting Started with Fastreid

Prepare pretrained model

If you use origin ResNet, you do not need to do anything. But if you want to use ResNet-ibn or ResNeSt, you need to download pretrain model in here. And then you need to put it in ~/.cache/torch/checkpoints or anywhere you like.

Then you should set the pretrain model path in configs/Base-bagtricks.yml.

Compile with cython to accelerate evalution

cd fastreid/evaluation/rank_cylib; make all

Training & Evaluation in Command Line

We provide a script in "tools/train_net.py", that is made to train all the configs provided in fastreid. 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, then run:

./tools/train_net.py --config-file ./configs/Market1501/bagtricks_R50.yml MODEL.DEVICE "cuda:0"

The configs are made for 1-GPU training.

If you want to train model with 4 GPUs, you can run:

./tools/train_net.py --config-file ./configs/Market1501/bagtricks_R50.yml --num-gpus 4

To evaluate a model's performance, use

./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.