Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/bwconrad/pricon
Official repository for "Pseudo-label Guided Joint Point Region Image-level Contrastive Learning for Task-specific Pretraining"
https://github.com/bwconrad/pricon
contrastive-learning pytorch pytorch-lightning semantic-segmentation
Last synced: about 1 month ago
JSON representation
Official repository for "Pseudo-label Guided Joint Point Region Image-level Contrastive Learning for Task-specific Pretraining"
- Host: GitHub
- URL: https://github.com/bwconrad/pricon
- Owner: bwconrad
- License: mit
- Created: 2022-07-22T12:16:46.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2022-12-09T18:30:06.000Z (about 2 years ago)
- Last Synced: 2024-05-14T00:18:10.547Z (8 months ago)
- Topics: contrastive-learning, pytorch, pytorch-lightning, semantic-segmentation
- Language: Python
- Homepage:
- Size: 5.34 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Awesome Lists containing this project
README
# Point Region Image-level Contrast
Official repository for ["Pseudo-label Guided Join Point Region Image-level Contrastive Learning for Task-specific Pretraining"](assets/thesis.pdf).
### Requirements
- Python 3.8+
- `pip install -r requirements.txt`
### Data Preparation
- Masks are expected to be single channel images at the same resolution as their corresponding image and consist of integer values between `[0, number-of-classes - 1]` where `number-of-classes` is the number of pseudo-label classes (including background).
- The [ `segmentation/` ](segmentation/) directory includes code to train a labeler network and [ `segmentation/scripts/generate_masks.py` ](segmentation/scripts/generate_masks.py) can be used to generate pseudo-label masks for a dataset using a trained model.
- Images and masks are expected to be in the correct order when the python `sort()` function is applied.
- i.e. Images and masks should have the same base name.
### Usage
#### _Training_
- To train with default settings run:
```
python train.py --gpus 1 --precision 16 --max_epoch 100 --data.image_dir path/to/images/
--data.mask_dir path/to/masks/ --data.n_classes --data.batch_size 256
```
- `` is the number of pseudo-label classes that the masks consist of.
- [`configs/`](configs/) contains example configuration files and can be run with `python train.py --config path/to/config/file`.
- Run `python train.py --help` for information on all options.
#### _Fine-tuning_
- To get the encoder weights from a pretraining checkpoint run:
```
python scripts/extract_encoder_weights.py --checkpoint path/to/pricon/checkpoint/
--output path/to/output/weights/
```
- The [`segmentation/`](segmentation/) directory contains training code for fine-tuning segmentation models using the extracted pretrained encoder weights.
#### _Cyclic Training_
After pretraining and fine-tuning the first time, generate new pseudo-label masks with the fine-tuned model by running:
```
python segmentation/scripts/generate_masks.py -w path/to/segmentation/checkpoint
-i path/to/unlabeled/images/
```
Next resume pretraining by running:
```
python train.py ... --model.resume_checkpoint path/to/pricon/checkpoint
--data.mask_dir path/to/new/masks/
```
where `path/to/pricon/checkpoint` is the checkpoint file from the previous pretraining run.