# phenopype-plugins
AI-plugins for [phenopype](https://www.phenopype.org/) - currently under development.
Currently, three plugin functions are available - all of them do image segmentation using pre-trained models:
- predict_fastSAM (Fast Segment Anything: https://docs.ultralytics.com/models/fast-sam/) - needs `ultralytics`
- predict_torch (Torchvision segmentation models: https://pytorch.org/vision/main/models.html) - needs `torch`
- predict_keras (Keras segmentation models https://keras.io/examples/vision/oxford_pets_image_segmentation/) - needs `keras`
## Installation
1\. Install phenopype (see https://www.phenopype.org/docs/installation/phenopype/ for more details):
pip install phenopype
2\. Install the plugins module:
pip install phenopype-plugins
3\. Install the dependencies
## Dependencies
If you have a GPU and the appropriate drivers install, make sure you install a fitting CUDA version first - e.g., v12.1:
pip install -c nvidia cuda-toolkit==12.1
### `torch`
1\. With GPU support:
pip install torch torchvision --index-url https://download.pytorch.org/whl/cu121
2\. Without GPU support:
pip install torch torchvision
### `keras`
1\. With GPU support:
pip install keras-gpu
2. Without GPU support:
pip install keras-gpu
### `ultralytics`
1\. Install Ultralytics BEFORE phenopype due to conflicting opencv-python (ultralytics) and opencv-contrib-python (phenopype) versions (see step 2 for alternatives):
pip install torch torchvision ## needed
pip install ultralytics
2\. If you have already installed phenopype and can't or don't want to uninstall it, you can do the following:
pip install ultralytics
## force reinstall opencv-contrib-python
pip install opencv-contrib-python==4.5.2.54 --force-reinstall
Raw data
{
"_id": null,
"home_page": null,
"name": "phenopype-plugins",
"maintainer": null,
"docs_url": null,
"requires_python": "==3.9.*",
"maintainer_email": null,
"keywords": "computer vision, biology",
"author": null,
"author_email": "Moritz L\u00fcrig <moritz.luerig@gmail.com>",
"download_url": "https://files.pythonhosted.org/packages/84/af/9b035ac06c73a48f7b110533d3a7143e8bacf3a196769396d57557ee59b3/phenopype_plugins-0.1.3.tar.gz",
"platform": null,
"description": "# phenopype-plugins\r\n\r\nAI-plugins for [phenopype](https://www.phenopype.org/) - currently under development. \r\n\r\nCurrently, three plugin functions are available - all of them do image segmentation using pre-trained models:\r\n\r\n - predict_fastSAM (Fast Segment Anything: https://docs.ultralytics.com/models/fast-sam/) - needs `ultralytics`\r\n - predict_torch (Torchvision segmentation models: https://pytorch.org/vision/main/models.html) - needs `torch`\r\n - predict_keras (Keras segmentation models https://keras.io/examples/vision/oxford_pets_image_segmentation/) - needs `keras`\r\n\r\n## Installation\r\n\r\n1\\. Install phenopype (see https://www.phenopype.org/docs/installation/phenopype/ for more details): \r\n\r\n pip install phenopype\r\n\r\n2\\. Install the plugins module:\r\n\r\n pip install phenopype-plugins\r\n\r\n3\\. Install the dependencies\r\n\r\n## Dependencies\r\n\r\nIf you have a GPU and the appropriate drivers install, make sure you install a fitting CUDA version first - e.g., v12.1:\r\n\r\n pip install -c nvidia cuda-toolkit==12.1\r\n\r\n### `torch`\r\n\r\n1\\. With GPU support:\r\n\r\n pip install torch torchvision --index-url https://download.pytorch.org/whl/cu121\r\n\r\n2\\. Without GPU support:\r\n\r\n pip install torch torchvision\r\n\r\n### `keras`\r\n\r\n1\\. With GPU support:\r\n\r\n pip install keras-gpu\r\n\r\n2. Without GPU support:\r\n\r\n pip install keras-gpu\r\n\r\n\r\n### `ultralytics`\r\n\r\n1\\. Install Ultralytics BEFORE phenopype due to conflicting opencv-python (ultralytics) and opencv-contrib-python (phenopype) versions (see step 2 for alternatives):\r\n\r\n \r\n pip install torch torchvision ## needed\r\n pip install ultralytics\r\n\r\n2\\. If you have already installed phenopype and can't or don't want to uninstall it, you can do the following:\r\n \r\n pip install ultralytics\r\n\r\n ## force reinstall opencv-contrib-python\r\n pip install opencv-contrib-python==4.5.2.54 --force-reinstall\r\n",
"bugtrack_url": null,
"license": null,
"summary": "Plugins for phenopype",
"version": "0.1.3",
"project_urls": {
"Bug Tracker": "https://github.com/phenopype/phenopype-plugins/issues",
"Homepage": "https://phenopype.org"
},
"split_keywords": [
"computer vision",
" biology"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "2bf16ae467bf1fbd2bdd346cb11299d1efddb57195894a6a38319d2d1d47e670",
"md5": "252acddb753b1313ef497d63e69733ff",
"sha256": "200536677517d9b96d23a09c61c967867747bb905e801022f5ed1e6d287cdd30"
},
"downloads": -1,
"filename": "phenopype_plugins-0.1.3-py3-none-any.whl",
"has_sig": false,
"md5_digest": "252acddb753b1313ef497d63e69733ff",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "==3.9.*",
"size": 12999,
"upload_time": "2024-05-17T03:02:01",
"upload_time_iso_8601": "2024-05-17T03:02:01.252221Z",
"url": "https://files.pythonhosted.org/packages/2b/f1/6ae467bf1fbd2bdd346cb11299d1efddb57195894a6a38319d2d1d47e670/phenopype_plugins-0.1.3-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "84af9b035ac06c73a48f7b110533d3a7143e8bacf3a196769396d57557ee59b3",
"md5": "f0a9deef55c3e33925b4bedd22ff281c",
"sha256": "4c9a2f5db998612b0ac67193b7cdcd38ce4b7fac9b029917bd8a429835fa6c9d"
},
"downloads": -1,
"filename": "phenopype_plugins-0.1.3.tar.gz",
"has_sig": false,
"md5_digest": "f0a9deef55c3e33925b4bedd22ff281c",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "==3.9.*",
"size": 12336,
"upload_time": "2024-05-17T03:02:02",
"upload_time_iso_8601": "2024-05-17T03:02:02.740671Z",
"url": "https://files.pythonhosted.org/packages/84/af/9b035ac06c73a48f7b110533d3a7143e8bacf3a196769396d57557ee59b3/phenopype_plugins-0.1.3.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-05-17 03:02:02",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "phenopype",
"github_project": "phenopype-plugins",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "phenopype-plugins"
}
JSON
