FSS-Codebase

This repo presents a neat and scalable codebase for few-shot segmentation research.

Major references: PFENet, CWT, and Segmenter.

Requisites

Test Env: Python 3.9.7 (Singularity)
- Path on NYU Greene: /scratch/hl3797/overlay-25GB-500K.ext3
Packages:
- torch (1.10.2+cu113), torchvision (0.11.3+cu113), timm (0.5.4)
- numpy, scipy, pandas, tensorboardX
- cv2, einops

Clone codebase

git clone https://github.com/hmdliu/FSS-Codebase && cd FSS-Codebase

Preparation

PASCAL-5i dataset

Note: Make sure the path in scripts/prepare_pascal.sh works for you.

# default data root: ../dataset/VOCdevkit/VOC2012
bash scripts/prepare_pascal.sh

COCO-20i dataset

You may refer to PFENet for more details.

Pretrained models

For ImageNet pre-trained models, please download it here (credits PFENet) and unzip as initmodel/. For models pre-trained on the base classes, you may find it here (credits CWT) and rename them as follows: pretrained/[dataset]/split[i]/pspnet_resnet[layers]/best.pth.

Dir explanations

initmodel: ImageNet pre-trained backbone weights. .pth
pretrained: Base classes pre-trained backbone weights. .pth
configs: Base configurations for experiments. .yaml
scripts: Training and helper scripts. .sh .slurm
results: Logs and checkpoints. .log .pth .yaml
src: Source code. .py

Sample Usage

exp_id aims to make efficient config modifications for experiment purposes. It follows the format of [exp_group]_[meta_cfg]_[train_cfg], see src/exp.py for a sample usage.

# debug mode (i.e., only log to shell)
python -m src.train --config configs/pascal_sample.yaml --exp_id sample_wd_pm11 --debug True

# submit to slurm
sbatch scripts/train_pascal.slurm configs/pascal_sample.yaml sample_wd_pm11

# output dir: results/sample/wd_pm11
tail results/sample/wd_pm11/output.log