sar-rarp50
Science Score: 26.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
○CITATION.cff file
-
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (8.2%) to scientific vocabulary
Repository
Basic Info
- Host: GitHub
- Owner: WANGCHAO0116
- License: apache-2.0
- Language: Jupyter Notebook
- Default Branch: main
- Size: 41 MB
Statistics
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
Machine Learning Task for SAR-RARP50
Environment Preparation
- create a virtual environment using conda and activate it.
shell conda create name openmmlab python=3.8 -y conda activate openmmlab - install PyTorch
shell pip install torch==1.12.0+cu116 torchvision==0.13.0+cu116 torchaudio==0.12.0 extra-index-url https://download.pytorch.org/whl/cu116 - install mmcv
shell pip install -U openmim mim install mmengine pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/cu116/torch1.12.0/index.html
Dataset
Raw Dataset:
Train set: https://rdr.ucl.ac.uk/articles/dataset/SAR-RARP50trainset/24932529
Test set: https://rdr.ucl.ac.uk/articles/dataset/SAR-RARP50testset/24932499
Data preprocess
A Python script is provided to extract frames from videos at 1 Hz.
shell
python sample_video.py -f <num_of_workers> -r <video_dir>
Preprocessed Dataset:
Download link: https://pan.baidu.com/s/1LK-ZTgA888DlipfscOoEkA?pwd=utsy
Please put the preprocessed dataset in the directory ./data.
The file structure of data should be as follows:
tree
data/
images/
video_01_000000000.png
video_01_000000060.png
...
video_50_000014520.png
labels/
video_01_000000000.png
video_01_000000060.png
...
video_50_000014520.png
splits/
train.txt
test.txt
Checkpoints:
Checkpoints are weights files (.pth).
Download link: https://drive.google.com/file/d/1FfTJg4Qu8N9Nrg2CH2Agmh-LKGjuc_CH/view?usp=sharing
Please put the checkpoints (.pth file) in the directory ./checkpoints
Train
Please refer to the file train.ipynb for the model training code.
Inference
We provide two python scripts to make inference for images and videos, respectively.
Make inference for image
shell
python image_inference.py --input_image <input_image_path> --model <model_checkpoint_path>
Make inference for video
shell
python video_inference.py --input_video <input_video_path> --model <model_checkpoint_path>
Owner
- Login: WANGCHAO0116
- Kind: user
- Repositories: 1
- Profile: https://github.com/WANGCHAO0116
GitHub Events
Total
Last Year
Dependencies
- pytorch/pytorch ${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel build
- pytorch/pytorch ${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel build
- cityscapesscripts *
- codecov *
- flake8 *
- interrogate *
- matplotlib *
- mmcls >=0.20.1
- mmcv-full <1.7.0,>=1.4.4
- numpy *
- packaging *
- prettytable *
- pytest *
- scipy *
- xdoctest >=0.10.0
- yapf *
- docutils ==0.16.0
- myst-parser *
- sphinx ==4.0.2
- sphinx_copybutton *
- sphinx_markdown_tables *
- mmcls >=0.20.1
- mmcv-full >=1.4.4,<1.7.0
- cityscapesscripts *
- mmcv *
- prettytable *
- scipy *
- torch *
- torchvision *
- matplotlib *
- mmcls >=0.20.1
- numpy *
- packaging *
- prettytable *
- scipy *
- codecov * test
- flake8 * test
- interrogate * test
- pytest * test
- xdoctest >=0.10.0 test
- yapf * test