View source for How to train your own AI model for an Inference Overlay

To train your own [https://pytorch.org/vision/main/models/mask_rcnn.html Mask R-CNN] AI model for an [[Inference Overlay]] in the {{software}}, you will need datasets to train on. 

To export these datasets using your projects in the {{software}}, follow this [[How to export AI Training Data|how-to]].

===Tygron AI Suite===
<onlyinclude>A Tygron AI Suite is available at [https://github.com/Tygron/tygron-ai-suite github]. This repository contains the necessary files to configure a Conda environment to train a new Mask R-CNN AI model.

[https://www.anaconda.com/products/navigator Anaconda Navigator] will be used to manage a Conda environment and run Jupyter Notebooks with python. 

[https://raw.githubusercontent.com/Tygron/tygron-ai-suite/refs/heads/main/example_config.ipynb Example notebooks] provided in the repository can be used as a basis to train your own model.

{{editor steps|title=train your own AI model for an [[Inference Overlay]]
| If you are not familiar with github, you can download the [https://github.com/Tygron/tygron-ai-suite/archive/refs/heads/main.zip zip] containing all the files that you need. Otherwise, clone the [https://github.com/Tygron/tygron-ai-suite git repository]
| Optionally unzip and open the folder containing the downloaded files; the local repository.
| [https://www.anaconda.com/products/navigator Download] and install ''Anaconda Navigator''.
| Open the ''Anaconda Navigator'' application
| Select the tab ''Environments'' and click on the Import button.
| Select the file "tygronai.yml" from the local repository of the tygron-ai-suite. This file will automatically setup the environment needed to open, edit and execute the ''Jupyter Notebooks''.
| Once configured, select the ''Home'' tab
| Open either the ''JupyterLab'' or ''Jupyter Notebook'' application. You might need to install it first by click on the install button.
| A browser will open with the selected application.
| Browse to the folder of the tygron-ai-suite repository and select the "example_config.ipynb".
| Adjust the following parameters: <pre>trainDirectory = "PATH TO TRAIN FILES"</pre><pre>testDirectory = "PATH TO TEST FILES"</pre> to the folders containing the exported datasets. See [[How to export an AI dataset|how to export AI datasets]] for more information.
| Press the double arrow button named ''Restart kernel and execute all cells'' to run the Jupyter Notebook. See the images below of what to expect.
| Eventually an [[ONNX]] file will be created.
}}

<gallery>
File:Inference_train_1.png
File:Inference_train_2.png
File:Inference_train_3.png
File:Inference_train_4.png
File:Inference_train_5.png
File:Inference_train_6.png
File:Inference_train_7.png
File:Inference_train_8.png
</gallery>

</onlyinclude>


{{article end
|seealso=* [[Demo Training Data Project]]
* [[Inference Overlay]]
* [[ONNX]]
* [[PyTorch]]
|notes=*The train sets can obtain data files from sub-folders. For example, if data was exported to ''user/documents/foliage/train/demo_training_data'', the '''PATH TO TRAIN FILES''' can be ''user/documents/foliage/train/''. This makes it easier to train on data exported from multiple projects and combining it with your own custom data.
*Once an [[ONNX]] file is created, open the folder containing this file and drag this file into the [[Tygron Client]] Application to automatically import it with an [[Inference Overlay]].
* In the step with the Configuration of the ''example_config'' notebook, ''cuda'' should be printed. If not, ''cpu'' will be printed. Training a model on your cpu is significantly slower. You might need to open the command line tool on your computer and run the command ''nvidia-smi'' to find out what version of cuda is supported on your gpu.
[[File:training_foliage_model_on_cuda.png]]
* To train a better model, increase the number of trained epochs, for example to 10:<pre>config.setEpochs(10)</pre> Please note that training with too many epochs may not improve the model; it will become too strict.
|howtos=*[[How to export AI Training Data]]}}