Training a custom AI model with Sensoft

Applies to

Sensoft Vision 2.4 and later

Process

Training a custom AI model to classify defects on filaments with Sensoft involves the following steps:

1. Choosing the classes
2. Labeling the photos
3. Training the model
4. Evaluate the model

It thus is a project of a few man-weeks to label the photos and evaluate the model and has costs, for Sensoptic to train and evaluate the model.  

Choosing the classes

Choose non-overlapping classes of defects, best between 6 and 12.  

Labeling

You can label photos for training directly in Sensoft.

Todo (how many)

Training the model

Todo

Evaluate the model

Todo