Overview
Napari-Easy-Augment-Batch-DL consists of a panel that interacts with several different Napari layers. The below screenshot shows Napari-Easy-Augment-Batch-DL open in Napari.
Main screen showing various layers and annotations.
Panel Groups
-
Load and Label: Load images then use the layers to label.
-
Augment: Apply augmentations to labels and save patches
-
Train and Predict: Use patches to train models, then use model to generate predictions.
Napari-Easy-Augment-Batch Layers
-
Images: The image set you are working with.
-
Labels: Manually drawn labels.
-
Predictions: Predictions from the trained model.
-
Label Box: Indicates which regions should be used for training. Manually drawn labels outside of a label box will not be used for training. In the figure above, the blue box is a label box.
-
ml_labels (experimental): Labels used for live ML training. Experiment with live machine learning training using these labels.
-
Object Box: Annotate individual objects with a bounding box. This type of annotation is used for training YOLO and other object detection methods.
-
Predicted Object Box: Predicted object box from YOLO or other object detection frameworks.