How to retrain simple MLP (palm_detection_full_inf_post_192x192.onnx) model with custom hand dataset ?

Question

How to retrain simple MLP (palm_detection_full_inf_post_192x192.onnx) model with custom hand dataset ?

Abhijeet241093 opened this issue 2 months ago · comments

Abhijeet241093 commented 2 months ago

Issue Type

Bug, Support

OS

Ubuntu

OS architecture

x86_64

Programming Language

Python

Framework

PyTorch, ONNX, TensorRT

Model name and Weights/Checkpoints URL

Model Name : palm_detection_full_inf_post_192x192.onnx

URL : https://github.com/PINTO0309/hand-gesture-recognition-using-onnx/tree/main/model/palm_detection

Description

How to retrain a simple MLP (palm_detection_full_inf_post_192x192.onnx) model with a custom hand dataset? Do you suggest any guidance or procedure to follow?

Relevant Log Output

Thank you Sir

URL or source code for simple inference testing code

https://github.com/PINTO0309/hand-gesture-recognition-using-onnx/tree/main/model/palm_detection

Katsuya Hyodo · Answer 1 · Mon Apr 22 2024 19:10:12 GMT+0800 (China Standard Time)

Ask Google's super engineers.

Abhijeet241093 · Answer 2 · Tue Apr 23 2024 10:08:29 GMT+0800 (China Standard Time)

Sorry, brother. Thank you for your direction.

However, unable to find any explanation for the same. If you can share logic or references about how to train an image with simple MLP? then it will be a great help.