How to Fine-Tuning the model on my custom dataset with same keypoints

[xddlj]

I would like to use the pretrained weights to initialize the model, then Fine-Tuning on my custom dataset with same keypoints.
I assume I would to do the following:

1. Set the configuration: PRETRAINED: 'premodels/LitePose-Auto-L.pth.tar' in mobile.yaml.
2. Set the LR: 0.0004 in mobile.yaml.
3. use the following script:
   python dist_train.py --cfg experiments/coco/mobilenet/mobile.yaml --superconfig mobile_configs/search-L.json DATASET.INPUT_SIZE 512 DATASET.OUTPUT_SIZE "[128, 256]"

If you have any procedures for Fine-Tuning on a custom dataset, I would appreciate it.