ValueError: Output tensors to a Model must be the output of a TensorFlow `Layer` (thus holding past layer metadata). Found: Tensor("Softmax:0", shape=(?, 88, 46), dtype=float32)

NinV / Korean-License-Plate-Recognition

Korean car license plate recognition using LPRNet

23 stars 3 forks source link

ValueError: Output tensors to a Model must be the output of a TensorFlow `Layer` (thus holding past layer metadata). Found: Tensor("Softmax:0", shape=(?, 88, 46), dtype=float32) #1

Closed Hedibouchelliga closed 4 years ago

Hedibouchelliga commented 4 years ago

Hello there can you please help me to overcome this error

I'm just trying to run the demo

NinV commented 4 years ago

Hi, I tested the code with TensorFlow 1.15 and 2.1.0 and It works fine. I can reproduce the error by using TensorFlow 1.13.1. So please check your TensorFlow version (should be 1.15 or 2). I also tested the code with Tensorflow 1.14 but I have a problem when training the model without the flag --save_weights_only. I hope this will help