# torch-kaleidoscope
A python library for providing PyTorch compatible colour representation utilities.
# weights
Pretrained weights for PyTorch networks are available from my research group where they are hosted and downloadable from the following format of URL:
`https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/<network>/<colour_space>.pth`
## available weights
- ResNet
- 50: [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet50/rgb.pth), [HSV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet50/hsv.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet50/h2sv.pth), [L\*a\*b\*](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet50/lab.pth), [L\*u\*v\*](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet50/luv.pth), [YUV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet50/yuv.pth)
- 101: [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet101/rgb.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet101/h2sv.pth)
- 152: [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet152/rgb.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet152/h2sv.pth)
- ResNeXt
- 50 32x4d: [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnext50_32x4d/rgb.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnext50_32x4d/h2sv.pth)
- 101 32x8d: [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnext101_32x8d/rgb.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnext101_32x8d/h2sv.pth)
- EfficientNetV2
- Small: [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/efficientnet_v2_s/rgb.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/efficientnet_v2_s/h2sv.pth)
- ConvNext
- Large: [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/convnext_large/rgb.pth), [HSV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/convnext_large/hsv.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/convnext_large/h2sv.pth), [H3SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/convnext_large/h3sv.pth), [YUV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/convnext_large/yuv.pth)
- Swin Transformer V2
- Tiny: : [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/swin_v2_t/rgb.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/swin_v2_t/h2sv.pth)
# acknowledgements
The work that lead to needing this, as well as huge swathes of the conversions code and testing present in this repository owe their existence to the scikit-image library[[1]](#1). In particular Nicolas Pinto, Ralf Gommers, Travis Oliphant, Matt Terry, Alex Izvorski and everyone else who contributed to the color module.
# references
<a id="1">[1]</a>
[scikit-image](https://doi.org/10.7717/peerj.453), the scikit-image team (2014)
Raw data
{
"_id": null,
"home_page": null,
"name": "torch-kaleidoscope",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": null,
"keywords": "machine learning, image processing, computer vision",
"author": null,
"author_email": "Alexis J S Payne <academia@aksan.dev>",
"download_url": "https://files.pythonhosted.org/packages/8e/b8/68da7ad41da18a23c409a55ed9245b16646020d6759376d8298d0ce87054/torch_kaleidoscope-0.1.1.tar.gz",
"platform": null,
"description": "# torch-kaleidoscope\r\nA python library for providing PyTorch compatible colour representation utilities.\r\n\r\n# weights\r\nPretrained weights for PyTorch networks are available from my research group where they are hosted and downloadable from the following format of URL:\r\n`https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/<network>/<colour_space>.pth`\r\n\r\n## available weights\r\n- ResNet\r\n - 50: [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet50/rgb.pth), [HSV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet50/hsv.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet50/h2sv.pth), [L\\*a\\*b\\*](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet50/lab.pth), [L\\*u\\*v\\*](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet50/luv.pth), [YUV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet50/yuv.pth)\r\n - 101: [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet101/rgb.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet101/h2sv.pth)\r\n - 152: [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet152/rgb.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnet152/h2sv.pth)\r\n- ResNeXt\r\n - 50 32x4d: [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnext50_32x4d/rgb.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnext50_32x4d/h2sv.pth)\r\n - 101 32x8d: [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnext101_32x8d/rgb.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/resnext101_32x8d/h2sv.pth)\r\n- EfficientNetV2\r\n - Small: [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/efficientnet_v2_s/rgb.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/efficientnet_v2_s/h2sv.pth)\r\n- ConvNext\r\n - Large: [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/convnext_large/rgb.pth), [HSV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/convnext_large/hsv.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/convnext_large/h2sv.pth), [H3SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/convnext_large/h3sv.pth), [YUV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/convnext_large/yuv.pth)\r\n- Swin Transformer V2\r\n - Tiny: : [RGB](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/swin_v2_t/rgb.pth), [H2SV](https://cvl.cs.nott.ac.uk/resources/kaleidoscope/backbones/swin_v2_t/h2sv.pth)\r\n\r\n\r\n\r\n\r\n# acknowledgements\r\nThe work that lead to needing this, as well as huge swathes of the conversions code and testing present in this repository owe their existence to the scikit-image library[[1]](#1). In particular Nicolas Pinto, Ralf Gommers, Travis Oliphant, Matt Terry, Alex Izvorski and everyone else who contributed to the color module. \r\n\r\n# references\r\n<a id=\"1\">[1]</a> \r\n[scikit-image](https://doi.org/10.7717/peerj.453), the scikit-image team (2014) \r\n",
"bugtrack_url": null,
"license": null,
"summary": "A tool kit for the use of non-RGB colour representations with torchvision and the PyTorch framework.",
"version": "0.1.1",
"project_urls": null,
"split_keywords": [
"machine learning",
" image processing",
" computer vision"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "340d38412b28a83eb8398531de7d6e64ba163c97e85cfcfe50b94e3a00c26a4c",
"md5": "5bbf6a425dd342590e711020e52636bc",
"sha256": "6450f08205877680edfdfa26f639321c74624eb37544ca664ecf78338c428b48"
},
"downloads": -1,
"filename": "torch_kaleidoscope-0.1.1-py3-none-any.whl",
"has_sig": false,
"md5_digest": "5bbf6a425dd342590e711020e52636bc",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 15159,
"upload_time": "2025-08-20T11:32:18",
"upload_time_iso_8601": "2025-08-20T11:32:18.262326Z",
"url": "https://files.pythonhosted.org/packages/34/0d/38412b28a83eb8398531de7d6e64ba163c97e85cfcfe50b94e3a00c26a4c/torch_kaleidoscope-0.1.1-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "8eb868da7ad41da18a23c409a55ed9245b16646020d6759376d8298d0ce87054",
"md5": "7f867414e4d430906d0e9e1e6c3e6a76",
"sha256": "ade0e9b1901fa5887f6d695b4668efe1ecdc3f2a60cd9e0d81f2c7bd5b7d5dfc"
},
"downloads": -1,
"filename": "torch_kaleidoscope-0.1.1.tar.gz",
"has_sig": false,
"md5_digest": "7f867414e4d430906d0e9e1e6c3e6a76",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 14463,
"upload_time": "2025-08-20T11:32:19",
"upload_time_iso_8601": "2025-08-20T11:32:19.541223Z",
"url": "https://files.pythonhosted.org/packages/8e/b8/68da7ad41da18a23c409a55ed9245b16646020d6759376d8298d0ce87054/torch_kaleidoscope-0.1.1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-08-20 11:32:19",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "torch-kaleidoscope"
}
JSON