SLUUDB7 August   2025 TMS320F28P550SJ

 

  1.   1
  2.   Abstract
  3.   Trademarks
  4. 1Introduction
  5. 2TMS320F28P55x
  6. 3Edge AI Studio
  7. 4Out-of-the-Box Demo (Smart Signal Classifier)
    1. 4.1 Dataset
      1. 4.1.1 Methods for Data Collection
        1. 4.1.1.1 Collection Process
      2. 4.1.2 Data Formats
    2. 4.2 Model Training
      1. 4.2.1 Preprocessing Options
    3. 4.3 Deploying to TI's Hardware
      1. 4.3.1 TVM Compiler
      2. 4.3.2 Model Execution
  8. 5Summary

Model Execution

  • The Deploy tab provides the option to export the compiled model for deploying onto the target device or exporting a trained model for further analysis by other engineering tools
     Deploy Tab Figure 4-16 Deploy Tab
  • The compiled model folder contains all the necessary configuration and code to run the model using the C2000 5.04 or later SDK
     Compiled Model
                            Folder Figure 4-17 Compiled Model Folder

Steps to import trained model into CCS project:

  1. Open Code Composer Studio and navigate to File-> import project ->.
    1. Select this folder

      \ti\c2000\C2000Ware_5_05_00_00\libraries\ai\feature_extract\c28\examples\ex_motor_fault_dataset_validation_f28p55x

  2. Import the necessary files from your compiled model download
    1. Unzip the compiled model
    2. Copy the artifacts folder into the imported project and overwriting the original files.
    3. Copy the test_vector.c and user_input config.h from the golden_vector directory inside the compiled model directory into your project. These overwrite the original test_vector and user_input_config.h file.
  3. Next, rebuild the project.
  4. Connect a F28P55 LaunchPad or control card. With the device debugger connected, press debug to verify the model feature extraction and neural network