Data

Tools

Indexes

Applications

Experiments

Open Source Science

https://github.com/cviu-csu/contrasive-yolo-world

https://github.com/cviu-csu/contrasive-yolo-world

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 (4.2%) to scientific vocabulary
Last synced: 9 months ago · JSON representation

Repository

Basic Info
  • Host: GitHub
  • Owner: CVIU-CSU
  • License: gpl-3.0
  • Language: Python
  • Default Branch: main
  • Size: 2.56 MB
Statistics
  • Stars: 0
  • Watchers: 0
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created 12 months ago · Last pushed 11 months ago
Metadata Files
Readme License

README.md

Raabin-det Dataset

This dataset is a refined detection-compatible version of Raabin-WBC. The images can be downloaded from (http://dl.raabindata.com/WBC/Secondmicroscope/Album4.zip), and the refined annotation files can be found in Contrastive-YOLO-World/data/Raabin-det/Raabin_det_test_instance.json.

Weights

The trained weights of Contrastive YOLO-World can be downloaded from (https://1drv.ms/u/c/752d9f191ce33dea/EXYBsTG0gJRInSmFBokmpg0B74Z1jP-xjwdtCupm5eW0cQ?e=a3ctep). Detailed training log can be seen at (https://1drv.ms/u/c/752d9f191ce33dea/Ef_5K5Kr3fdHp-U79eGMlK0BAGWm2FvrnjcrRmc2QMn8iA?e=qutf8g).

Training

You can train the model using the following code:

bash export PYTHONPATH="/root/Contrastive-YOLO-World:$PYTHONPATH" bash tools/dist_train.sh yolo_world_l_dual_vlpan_2e-4_80e_8gpus_finetune_coco.py 8

Evaluation

You can also evaluate the model using our provided weights, you can run the code:

bash export PYTHONPATH="/root/Contrastive-YOLO-World:$PYTHONPATH" bash tools/dist_test.sh yolo_world_l_dual_vlpan_2e-4_80e_8gpus_finetune_coco.py contrastive-yolo-world_3e-4.pth 8

Inference

You can perform inference using the following code:

bash export PYTHONPATH="/root/Contrastive-YOLO-World:$PYTHONPATH" python demo/image_demo.py yolo_world_l_dual_vlpan_2e-4_80e_8gpus_finetune_coco.py contrastive-yolo-world_3e-4.pth images text_path.txt --output-dir output_path

Owner

  • Name: CVIU-CSU
  • Login: CVIU-CSU
  • Kind: organization

GitHub Events

Total
  • Push event: 4
Last Year
  • Push event: 4

Dependencies

Dockerfile docker
  • nvidia/cuda 11.8.0-devel-ubuntu22.04 build
deploy/easydeploy/examples/requirements.txt pypi
  • onnxruntime *
  • opencv-python ==4.7.0.72
pyproject.toml pypi
  • mmcv *
  • mmcv-lite >=2.0.0rc4
  • mmdet ==3.0.0
  • mmengine >=0.7.1
  • mmyolo @ git+https://github.com/onuralpszr/mmyolo.git
  • numpy *
  • opencv-python *
  • openmim *
  • supervision ==0.19.0
  • tokenizers *
  • torch >=1.11.0
  • torchvision >=0.16.2
  • transformers *
  • wheel *
requirements/basic_requirements.txt pypi
  • albumentations *
  • mmcv ==2.0.0
  • mmdet ==3.0.0
  • mmengine ==0.10.3
  • mmyolo ==0.6.0
  • opencv-python ==4.9.0.80
  • opencv-python-headless ==4.2.0.34
  • timm ==0.6.13
  • transformers ==4.36.2
requirements/demo_requirements.txt pypi
  • gradio ==4.16.0
  • supervision *
requirements/onnx_requirements.txt pypi
  • onnx *
  • onnxruntime *
  • onnxsim *
  • supervision *