search

PINTO0309 / PINTO_model_zoo

A repository for storing models that have been inter-converted between various frameworks. Supported frameworks are TensorFlow, PyTorch, ONNX, OpenVINO, TFJS, TFTRT, TensorFlowLite (Float32/16/INT8), EdgeTPU, CoreML.
https://qiita.com/PINTO
MIT License
3.59k stars 573 forks source link

[FaceMesh-with-Attention] Retraining capability #270

Closed dannyway03 closed 2 years ago

dannyway03 commented 2 years ago

Issue Type

Support

OS

Ubuntu

OS architecture

x86_64

Programming Language

C++

Framework

TensorFlowLite

Model name and Weights/Checkpoints URL

FaceMesh-with-Attention

Description

Hi,

given the saved_model folder, would it be possible to re-train or continue the training of the converted model? In particular, I'm interested in FaceMesh-with-Attention, but I'm not really an expert on this topic.

many thanks. D-

Relevant Log Output

No response

URL or source code for simple inference testing code

No response

PINTO0309 commented 2 years ago

given the saved_model folder, would it be possible to re-train or continue the training of the converted model?

No. You will need to rebuild the model using TensorFlow or PyTorch. https://google.github.io/mediapipe/solutions/face_mesh.html

PINTO0309 commented 2 years ago

In any case, you should consider whether FaceMesh-with-Attention is really the best solution for you.