Use MATLAB to Train on Jetson Nano: Unterschied zwischen den Versionen
Zur Navigation springen
Zur Suche springen
Keine Bearbeitungszusammenfassung |
Keine Bearbeitungszusammenfassung |
||
| Zeile 8: | Zeile 8: | ||
#The fundamental of training an AI system is called deep learning. MATLAB provides a variety of helpful toolboxes, pre-trained networks, and tools to process data into useful input for an AI [https://de.mathworks.com/videos/search.html?q=&fq%5B%5D=product:GC&page=1]. | #The fundamental of training an AI system is called deep learning. MATLAB provides a variety of helpful toolboxes, pre-trained networks, and tools to process data into useful input for an AI [https://de.mathworks.com/videos/search.html?q=&fq%5B%5D=product:GC&page=1]. | ||
#The training can be stored as a model, usually in a .MAT format, which can be reused. Hence, we do not have to train a model every time we want to use it. | #The training can be stored as a model, usually in a .MAT format, which can be reused. Hence, we do not have to train a model every time we want to use it. | ||
#Data preprocessing—including image resizing, augmentation, and normalization—is handled with MATLAB functions such as imageDatastore and augmentedImageDatastore. | |||
{| class="wikitable" | |||
|+ Tabelle 4: Deep Learning with MATLAB Setup | |||
|- | |||
! # !! !! Description !! Pictures | |||
|- | |||
| 1 || Access GPU APP || || [[Datei:Gpu coder APP.png|mini]] | |||
|- | |||
| 2 || Select Targeted GPU|| | |||
|- | |||
| 3 || Generate c/c++ || || | |||
|} | |||
*Gpu Coder | *Gpu Coder | ||
#GPU Coder generates optimized CUDA® C++ from your MATLAB® algorithms, unlocking the parallel compute power of the Jetson GPU for real-time processing on your JetRacer and improve 40 times the performance on the jetracer [https://www.mathworks.com/help/coder/nvidia.html?utm_source=chatgpt.com][https://www.youtube.com/watch?v=emVRH4HfhQY&list=PLRcFXSK3rd79jpZLqttVDlBGrl6cYK3iS]. | #GPU Coder generates optimized CUDA® C++ from your MATLAB® algorithms, unlocking the parallel compute power of the Jetson GPU for real-time processing on your JetRacer and improve 40 times the performance on the jetracer [https://www.mathworks.com/help/coder/nvidia.html?utm_source=chatgpt.com][https://www.youtube.com/watch?v=emVRH4HfhQY&list=PLRcFXSK3rd79jpZLqttVDlBGrl6cYK3iS]. | ||
#Helps target GPUs for automotive applications. | #Helps target GPUs for automotive applications. | ||
{| class="wikitable" | |||
|+ Tabelle 4: GPU Coder Setup | |||
|- | |||
! # !! !! Description !! Pictures | |||
|- | |||
| 1 || Access GPU APP || || [[Datei:Gpu coder APP.png|mini]] | |||
|- | |||
| 2 || Select Targeted GPU|| | |||
|- | |||
| 3 || Generate c/c++ || || | |||
|} | |||
=== [[Result]] === | === [[Result]] === | ||
Version vom 1. Juni 2025, 14:01 Uhr
Setup
Overview
To train the AI using MATLAB, I will write a script on MATLAB and then use GPU Coder to flash the function/model on the JetRacer. So all the debugging process will be on MATLAB.
Key Concepts
- Deep Learning with MATLAB
- The fundamental of training an AI system is called deep learning. MATLAB provides a variety of helpful toolboxes, pre-trained networks, and tools to process data into useful input for an AI [1].
- The training can be stored as a model, usually in a .MAT format, which can be reused. Hence, we do not have to train a model every time we want to use it.
- Data preprocessing—including image resizing, augmentation, and normalization—is handled with MATLAB functions such as imageDatastore and augmentedImageDatastore.
| # | Description | Pictures | |
|---|---|---|---|
| 1 | Access GPU APP |  | |
| 2 | Select Targeted GPU | ||
| 3 | Generate c/c++ |
- Gpu Coder
- GPU Coder generates optimized CUDA® C++ from your MATLAB® algorithms, unlocking the parallel compute power of the Jetson GPU for real-time processing on your JetRacer and improve 40 times the performance on the jetracer [2][3].
- Helps target GPUs for automotive applications.
| # | Description | Pictures | |
|---|---|---|---|
| 1 | Access GPU APP | | |
| 2 | Select Targeted GPU | ||
| 3 | Generate c/c++ |