If in deep learning it is necessary to first use instance segmentation to segment objects, and then use image classification to distinguish positive and negative, how does such a model cascade operate? Is there any documentation?
- After the instance segmentation model training is completed, click Verify;
- Add an image classification module and import data from the previous level;
- After the image classification model training is completed, click Export.
The document can be viewed in docs: “Help - User Manual - Meck-DLK Application Guide - Cascading Module”