VisualizationHook

class mmcls.engine.hooks.VisualizationHook(enable=False, interval=5000, show=False, out_dir=None, **kwargs)

Classification Visualization Hook. Used to visualize validation and testing prediction results.

• If out_dir is specified, all storage backends are ignored and save the image to the out_dir.

• If show is True, plot the result image in a window, please confirm you are able to access the graphical interface.

Parameters

• enable (bool) – Whether to enable this hook. Defaults to False.

• interval (int) – The interval of samples to visualize. Defaults to 5000.

• show (bool) – Whether to display the drawn image. Defaults to False.

• out_dir (str, optional) – directory where painted images will be saved in the testing process. If None, handle with the backends of the visualizer. Defaults to None.

• **kwargs – other keyword arguments of mmcls.visualization.ClsVisualizer.add_datasample().

after_test_iter(runner, batch_idx, data_batch, outputs)

Visualize every self.interval samples during test.

Parameters

• runner (Runner) – The runner of the testing process.

• batch_idx (int) – The index of the current batch in the test loop.

• data_batch (dict) – Data from dataloader.

• outputs (Sequence[DetDataSample]) – Outputs from model.

after_val_iter(runner, batch_idx, data_batch, outputs)

Visualize every self.interval samples during validation.

Parameters

• runner (Runner) – The runner of the validation process.

• batch_idx (int) – The index of the current batch in the val loop.

• data_batch (dict) – Data from dataloader.

• outputs (Sequence[ClsDataSample]) – Outputs from model.