Tiago Freitas

I got the same error with a compute module V3 with dual cameras. Your suggestion resolved my problem thanks. When I analyze the RPI_REVISION it outputs the RPI3 model instead...

That seems a problem with keras ImageDataGenerator. Can you share the load dataset code if you changed it? How is organized your image directory path?

Glad you found it out!

Did you edit the loader? What TensorFlow version are you using? It seems related to ImageDataGenerator

This code was supposed to work with tensorflow 1. I am not sure that it works with tf2. Like @kunalr326 referred, it could be related to tf version.

Hi @Abdulrahman-CS99 This code was written in tensorflow 1. You might need to re-adapt it to Tensorflow V2

Hi @satyam2369, Without a look at the changes you made to the code I cannot help much. Notice that the code was written in Tensorflow 1, so you will need...

Hi @iamshahd, At the time I was not using a good GPU so I was not able to get the exact results as I stated in the docs: `The results...

Hi! After having a model trained, you can easily replicate the prediction in a new dataset as in the validation part of the code.