getting_started.md

Getting Started

Data Preparation

Download the data (VOC, Cityscapes) and pre-trained models:

DATA/
|-- city
|-- pascal_voc
|-- pytorch-weight
|   |-- resnet50_v1c.pth
|   |-- resnet101_v1c.pth

For CPS+FPL (e.g., VOC dataset)

$ cd ./FPL_based_on_CPS/exp.voc/voc.res101v3+.CPS+CutMix+FPL
$ bash script_FPL.sh

You may need:

• Adapt data and model path in config_FPL.py.
• Specify some variables(e.g., the path to your snapshot dir) in script_FPL.sh.
• Change C.labeled_ratio in config_FPL.py for other data partitions.
• The core contribution of FPL is the topk_ce_FPL function in train_FPL.py.

For AEL+FPL (e.g., Cityscapes dataset)

$ cd ./FPL_based_on_AEL/experiments/cityscapes_2
$ bash train.sh

You may need:

• Adapt data path in config.yaml.
• Adapt model path in ./FPL_based_on_AEL/semseg/models/resnet.py.
• Change n_sup in config.yaml for other data partitions.
• The core contribution of FPL is the topk_ce_FPL function in ./FPL_based_on_AEL/semseg/utils/loss_helper_topk.py.