...
PEKAT VISION software offers training of models using both Supervised and Unsupervised methods. Detailed modules information can be found within the article /wiki/spaces/KB3141/pages/533528878 Modules.
The Unsupervised module available is /wiki/spaces/KB3141/pages/533529141 Anomaly Detector,where the only required end-user input is to assign at least 30% of the images to the OK class.
...
The Training mode for PEKAT is available for the following AI Modules:
/wiki/spaces/KB3141/pages/533529141
/wiki/spaces/KB3141/pages/533529197
/wiki/spaces/KB3141/pages/533529217
/wiki/spaces/KB3141/pages/533529284
/wiki/spaces/KB3141/pages/533529386Anomaly Detector
Below you can find a Training Overview Flowchart, which demonstrates the overall training steps. However, for detailed training information about a specific module, please access the desired module’s page.
...
The Training begins after clicking on the Start Training button, from that point a dialog window will display a training chart only for the following modules:
/wiki/spaces/KB3141/pages/533529197
/wiki/spaces/KB3141/pages/533529217
/wiki/spaces/KB3141/pages/533529284 Classifier
Based on the chart displayed, its graph gives you an indication of the ideal moment to stop the training.
...
Please note the chart is not available for the Module /wiki/spaces/KB3141/pages/533529141 Anomaly Detector, since by default we have the following:
...
As described above, for both functions, the number of epochs is determined before training starts, therefore loss function graph is not available.
For the /wiki/spaces/KB3141/pages/533529386 OCR module, the number of epochs is also set before starting the training.
...
Extend model function is only available for the modules /wiki/spaces/KB3141/pages/533529197, /wiki/spaces/KB3141/pages/533529217, and /wiki/spaces/KB3141/pages/533529386 Classifier, Detector, and OCR.
This function allows you to perform further training on an already-trained model.
...
For a detailed explanation regarding augmentation, visit the article /wiki/spaces/KB3141/pages/533463892 Augmentation Glossary.
Model Validation
After training, the new model is automatically selected and its preview is shown. A different model can be selected from the model list using the radio Change button and then clicking on the Preview button desired model to check the model’s results, as shown in the picture pictures below. When a model is selected, the statistics will then be based on the selected model.
It is also possible to select multiple models by pressing CTRL and clicking on the models you want to select.
...
Fast Training Mode & Deep Learning Mode
At first, we recommend running different models in Fast Training mode, using different sizes of view-finder and training parameters (brightness resistance, resistance to deviation, etc.) to find out what combination works best in your case. When you find suitable settings, but you would like to achieve even more precise results, you may try Deep Training mode, with more training cycles, however, using the same training parameters (settings).
To delete the trained model click the trash icon on the right or use CTRL for selecting the model and then click the DELETE SELECTED button.
...
Information on the training process and conditions used for a particular model are available by clicking on the info icon in the list of models, as shown in the picture below.
...