Skip to content

Latest commit

 

History

History
20 lines (16 loc) · 1.58 KB

File metadata and controls

20 lines (16 loc) · 1.58 KB

Deployment

Overview of directories and files

Command to perform the deployment

After training and quantization all the required files, stats, and values are saved in the /training_quantization/ folder.
Therefore, during the execution, the /training_quantization/deep_learning_config.ini is read to retrieve useful information, such as quantization values and versions of the models.
Please keep in mind that if you want to perform the deployment on older trained versions of the models, you would have to modify the QUANTIZATION_NAVIGATION and QUANTIZATION_CLASSIFICATION parameters in the file accordingly.
You can find the inputs and output quantization values in the folders on which your models are saved, as a .txt file.

To perform the deployment, execute the following command: python3 main_deployment.py

Files saved throughout the execution

During the execution, two new directories are created and filled with the respective deployed models in C code.

  • classification_model_quant/
  • navigation_model_quant/

Manual steps to join the two deployed CNNs scripts produced in this execution