# NRTK EXPLORER
NRTK Explorer is a web application for exploring image datasets. It provides
insights of a image dataset in [COCO][3] format and it evaluate image
transformation and perturbation resilience of object recognition DL models. It
is built using [trame][1] by the [kitware][2] team.
## Features
- Explore image datasets in COCO format.
- Apply parametrized image degradation (such as blur) to the images.
- Benchmark dataset resilience with a differential PCA|UMAP analysis on the
embeddings of the images and their transformations.
- Evaluate object detection DL models in both the source images and its
transformations.
- When possible it will attempt to utilize the user GPU as much as possible to
speedup its computations.
## Installing
Install it from pypi:
```bash
pip install nrtk-explorer
```
## Usage
Explore Hugging Face hosted [dataset](https://huggingface.co/datasets/rafaelpadilla/coco2017):
```bash
nrtk-explorer --dataset rafaelpadilla/coco2017
```
Compare inference results for Hugging Face hosted models:
```bash
nrtk-explorer --dataset cppe-5 --models qubvel-hf/detr-resnet-50-finetuned-10k-cppe5 ashaduzzaman/detr_finetuned_cppe5
```
2 COCO format datasets are available at: https://github.com/vicentebolea/nrtk_explorer_datasets/
```bash
git clone https://github.com/vicentebolea/nrtk_explorer_datasets.git
nrtk-explorer --dataset ./nrtk_explorer_datasets/coco-od-2017/mini_val2017.json ./nrtk_explorer_datasets/OIRDS_v1_0/oirds.json
```
## CLI flags and options
- `--dataset` specify the path to a [COCO dataset](https://roboflow.com/formats/coco-json) JSON file,
a [Hugging Face dataset](https://huggingface.co/datasets?task_categories=task_categories:object-detection) repository name,
or a directory loadable by the [Dataset](https://huggingface.co/docs/datasets/index) library.
You can specify multiple datasets using a space as the
separator. Example: `nrtk-explorer --dataset ../foo-dir/coco.json cppe-5`
- `--repository` Specify an existing directory where exported datasets will be saved to and loaded from.
- `--download` Cache Hugging Face Hub datasets locally instead of streaming them.
When datasets are streamed, nrtk-explorer limits the number of loaded images.
- `--models` specify the Hugging Face Hub [object detection](https://huggingface.co/models?pipeline_tag=object-detection&library=transformers&sort=trending)
repository name or a directory loadable by the [Transformers](https://huggingface.co/docs/transformers/index) library. Load multiple models using space as the separator.
Example: `nrtk-explorer --models hustvl/yolos-tiny facebook/detr-resnet-50`
- `-h|--help` show the help for the command line options. nrtk-explorer inherits the trame
command line options and flags.
## Contribute to NRTK_EXPLORER
```bash
git clone https://github.com/Kitware/nrtk-explorer.git
cd nrtk-explorer
python3 -m venv .venv
source .venv/bin/activate
pip install -U pip
pip install -e '.[dev]'
pytest .
```
For more details on setting up a development environment see [DEVELOPMENT docs](docs/source/manual/DEVELOPMENT.rst).
### Create release
1. Merge `main` to `release` with a _merge commit_.
2. Run "Create Release" workflow with workflow from `release` branch.
3. Merge `release` to `main` with a _merge commit_.
4. Check package versions in Conda Feedstock [meta.yaml file](https://github.com/conda-forge/nrtk-explorer-feedstock/blob/main/recipe/meta.yaml)
[1]: https://trame.readthedocs.io/en/latest/
[2]: https://www.kitware.com/
[3]: https://cocodataset.org/
Raw data
{
"_id": null,
"home_page": null,
"name": "nrtk-explorer",
"maintainer": null,
"docs_url": null,
"requires_python": "<3.13,>=3.10",
"maintainer_email": null,
"keywords": "Application, Framework, Interactive, Python, Web",
"author": null,
"author_email": "Alessandro Genova <alessandro.genova@kitware.com>, Vicente Adolfo Bolea Sanchez <vicente.bolea@kitware.com>",
"download_url": "https://files.pythonhosted.org/packages/df/a0/9be2d1b6d0a545017577c7ebff70f8536a8e27a604974b2f9ef3ab6c17ec/nrtk_explorer-0.8.3.tar.gz",
"platform": null,
"description": "# NRTK EXPLORER\n\nNRTK Explorer is a web application for exploring image datasets. It provides\ninsights of a image dataset in [COCO][3] format and it evaluate image\ntransformation and perturbation resilience of object recognition DL models. It\nis built using [trame][1] by the [kitware][2] team.\n\n\n\n## Features\n\n- Explore image datasets in COCO format.\n- Apply parametrized image degradation (such as blur) to the images.\n- Benchmark dataset resilience with a differential PCA|UMAP analysis on the\n embeddings of the images and their transformations.\n- Evaluate object detection DL models in both the source images and its\n transformations.\n- When possible it will attempt to utilize the user GPU as much as possible to\n speedup its computations.\n\n## Installing\n\nInstall it from pypi:\n\n```bash\npip install nrtk-explorer\n```\n\n## Usage\n\nExplore Hugging Face hosted [dataset](https://huggingface.co/datasets/rafaelpadilla/coco2017):\n\n```bash\nnrtk-explorer --dataset rafaelpadilla/coco2017\n```\n\nCompare inference results for Hugging Face hosted models:\n\n```bash\nnrtk-explorer --dataset cppe-5 --models qubvel-hf/detr-resnet-50-finetuned-10k-cppe5 ashaduzzaman/detr_finetuned_cppe5\n```\n\n2 COCO format datasets are available at: https://github.com/vicentebolea/nrtk_explorer_datasets/\n\n```bash\ngit clone https://github.com/vicentebolea/nrtk_explorer_datasets.git\nnrtk-explorer --dataset ./nrtk_explorer_datasets/coco-od-2017/mini_val2017.json ./nrtk_explorer_datasets/OIRDS_v1_0/oirds.json\n```\n\n## CLI flags and options\n\n- `--dataset` specify the path to a [COCO dataset](https://roboflow.com/formats/coco-json) JSON file,\n a [Hugging Face dataset](https://huggingface.co/datasets?task_categories=task_categories:object-detection) repository name,\n or a directory loadable by the [Dataset](https://huggingface.co/docs/datasets/index) library.\n You can specify multiple datasets using a space as the\n separator. Example: `nrtk-explorer --dataset ../foo-dir/coco.json cppe-5`\n- `--repository` Specify an existing directory where exported datasets will be saved to and loaded from.\n- `--download` Cache Hugging Face Hub datasets locally instead of streaming them.\n When datasets are streamed, nrtk-explorer limits the number of loaded images.\n- `--models` specify the Hugging Face Hub [object detection](https://huggingface.co/models?pipeline_tag=object-detection&library=transformers&sort=trending)\n repository name or a directory loadable by the [Transformers](https://huggingface.co/docs/transformers/index) library. Load multiple models using space as the separator. \n Example: `nrtk-explorer --models hustvl/yolos-tiny facebook/detr-resnet-50`\n- `-h|--help` show the help for the command line options. nrtk-explorer inherits the trame\n command line options and flags.\n\n\n\n## Contribute to NRTK_EXPLORER\n\n```bash\ngit clone https://github.com/Kitware/nrtk-explorer.git\ncd nrtk-explorer\npython3 -m venv .venv\nsource .venv/bin/activate\npip install -U pip\npip install -e '.[dev]'\npytest .\n```\n\nFor more details on setting up a development environment see [DEVELOPMENT docs](docs/source/manual/DEVELOPMENT.rst).\n\n### Create release\n\n1. Merge `main` to `release` with a _merge commit_.\n2. Run \"Create Release\" workflow with workflow from `release` branch.\n3. Merge `release` to `main` with a _merge commit_.\n4. Check package versions in Conda Feedstock [meta.yaml file](https://github.com/conda-forge/nrtk-explorer-feedstock/blob/main/recipe/meta.yaml)\n\n[1]: https://trame.readthedocs.io/en/latest/\n[2]: https://www.kitware.com/\n[3]: https://cocodataset.org/\n",
"bugtrack_url": null,
"license": null,
"summary": "Model Visualizer",
"version": "0.8.3",
"project_urls": null,
"split_keywords": [
"application",
" framework",
" interactive",
" python",
" web"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "fac57d1cc5639b6f90dab801fa974e2b4ed8e1d0b80ff68b21095e7ff95b781c",
"md5": "8c8ef65f3ca0c95ead3a5577d18c8ace",
"sha256": "514b7c18df8fb616e26367c9e5f25325ab5252375204eaae8835b91daca1ac9e"
},
"downloads": -1,
"filename": "nrtk_explorer-0.8.3-py3-none-any.whl",
"has_sig": false,
"md5_digest": "8c8ef65f3ca0c95ead3a5577d18c8ace",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<3.13,>=3.10",
"size": 3529128,
"upload_time": "2025-10-24T20:31:00",
"upload_time_iso_8601": "2025-10-24T20:31:00.479911Z",
"url": "https://files.pythonhosted.org/packages/fa/c5/7d1cc5639b6f90dab801fa974e2b4ed8e1d0b80ff68b21095e7ff95b781c/nrtk_explorer-0.8.3-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "dfa09be2d1b6d0a545017577c7ebff70f8536a8e27a604974b2f9ef3ab6c17ec",
"md5": "08460f441600e5cc6afd59f850f47c84",
"sha256": "a0dff245866417896e0948edd021300adf9d826d66dbbdce6b17a07ec720211f"
},
"downloads": -1,
"filename": "nrtk_explorer-0.8.3.tar.gz",
"has_sig": false,
"md5_digest": "08460f441600e5cc6afd59f850f47c84",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<3.13,>=3.10",
"size": 4846497,
"upload_time": "2025-10-24T20:31:02",
"upload_time_iso_8601": "2025-10-24T20:31:02.126857Z",
"url": "https://files.pythonhosted.org/packages/df/a0/9be2d1b6d0a545017577c7ebff70f8536a8e27a604974b2f9ef3ab6c17ec/nrtk_explorer-0.8.3.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-10-24 20:31:02",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "nrtk-explorer"
}
JSON