Metadata-Version: 2.1
Name: icevision_dashboards
Version: 0.0.5
Summary: Provides dashboard components and dashboards to get insight into data.
Home-page: https://github.com/fstroth/icevision_dashboards/tree/master/
Author: Frederik Strothmann
Author-email: frstrothmann@gmail.com
License: Apache Software License 2.0
Description: # IceVisionDashboard
        
        
        
        IceVisionDashboard is an extension to the [IceVision](https://github.com/airctic/icevision) object detection framework. This extension provides different `dashboards` to investigate datasets, create new datasets and analyse the results of a training.
        
        The libary provides dashboards to inspect datasets, create new datasets, investigate the results of a training and much more.
        
        ## Examples
        
        ```python
        from icevision_dashboards.data import BboxRecordDataset
        from icevision_dashboards.dashboards import ObjectDetectionDatasetOverview
        
        # load some data from the icedata
        data_dir = icedata.fridge.load_data()
        class_map = icedata.fridge.class_map()
        parser = icedata.fridge.parser(data_dir)
        train_records, valid_records = test_parser.parse()
        
        # create a dataset that can be consumed by the dashboards
        train_dash_ds = BboxRecordDataset(train_records, class_map)
        
        # create a new dashboard instance and display it with the .show() function
        overview_dashboard = ObjectDetectionDatasetOverview(train_dash_ds, width=1500, height=900)
        overview_dashboard.show()
        ```
        
        The output will look like this:
        
        ![dataset_overivew_0](imgs/dataset_overview_0.png)
        ![dataset_overivew_1](imgs/dataset_overview_1.png)
        <div style="text-align:center"><img src="imgs/dataset_overview_2.png" /></div>
        
        ![dataset_overivew_3](imgs/dataset_overview_3.png)
        
        ## Contributing
        
        If you want to contribute add the following lines to your `pre-commit` file to ensure the notebook cell output don't get pushed into the repo.
        
        ```bash
        # ensure the oupt of the notebooks is empty
        jupyter nbconvert --ClearOutputPreprocessor.enabled=True --inplace nbs/*.ipynb
        git add .
        ```
        
Keywords: machine learning,dashboard,deep learning
Platform: UNKNOWN
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Natural Language :: English
Classifier: Programming Language :: Python :: 3.8
Requires-Python: >=3.8
Description-Content-Type: text/markdown
