Menu: Tools / Deep learning AI / Classify
Here you will be able to classify your projects automatically by using a trained model. In the dropbox you will select the model you want to use for the classification. Once the model selected, the classes availables for classification will be displayed.
If for some reason, the list displayed does not match your current classes, you can simply click 
. Then the classes in the explorer window will be adjust to the one of your model.
In option, you can select between:
- GPU mode : Will use your graphic card to do the classification
- CPU mode : Will use your processor to do the classification
- Number of partitions : This let you split your project in multiple area. Important to use for large project (large surface not points).
- Number of votes batch:This let you choose the performance of the AI classification. The lower the faster, the higher the more precise (sweet spot is between 10 and 20).
- This option defines how many samples will be used per process iteration. Higher the number, more memory will be used. In the case that you run out of memory, reduce this number.
Once the parameters are setup just click 
to start the process.
Note : Using CPU will take longer than the GPU. So far the GPU mode only support NVidia video cards.
