2020-10-01 18:10:02 +08:00
|
|
|
# Hyper-Parameter Optimization in FastReID
|
|
|
|
|
|
|
|
|
|
This project includes training reid models with hyper-parameter optimization.
|
|
|
|
|
|
|
|
|
|
Install the following
|
|
|
|
|
|
|
|
|
|
```bash
|
|
|
|
|
pip install 'ray[tune]'
|
2020-10-10 17:35:25 +08:00
|
|
|
pip install hpbandster ConfigSpace hyperopt
|
2020-10-01 18:10:02 +08:00
|
|
|
```
|
|
|
|
|
|
2020-10-10 17:35:25 +08:00
|
|
|
## Example
|
2020-10-01 18:10:02 +08:00
|
|
|
|
2020-10-10 17:35:25 +08:00
|
|
|
This is an example for tuning `batch_size` and `num_instance` automatically.
|
|
|
|
|
|
|
|
|
|
To train hyperparameter optimization with BOHB(Bayesian Optimization with HyperBand) search algorithm, run
|
2020-10-01 18:10:02 +08:00
|
|
|
|
|
|
|
|
```bash
|
2020-10-10 17:35:25 +08:00
|
|
|
python3 projects/HPOReID/train_hpo.py --config-file projects/HPOReID/configs/baseline.yml --srch-algo "bohb"
|
2020-10-01 18:10:02 +08:00
|
|
|
```
|
|
|
|
|
|
|
|
|
|
## Known issues
|
|
|
|
|
todo
|
