Deep GAN Training and results

abhishek13yadv · April 16, 2020, 5:10am

I have got an error in running the code
=>
https://colab.research.google.com/drive/1J3hhLXxaYONWqQy7nmNPYa5tzVlrHaUu

Error : expected conv2d_7_input to have shape (28, 28, 1) but got array with shape (28, 28, 64)

S18CRX0120 · April 16, 2020, 5:36am

Hey @abhishek13yadv, your generator model is complete, it should be like that, one having output shape (28,28,1). Add this last layer in generator model,
generator.add(Conv2D(1, kernel_size=(5, 5), padding=‘same’, activation=‘tanh’))
Also you can take reference from here,

github.com

coding-blocks-archives/machine-learning-online-2018/blob/master/Generative Adversarial Networks (GANs)/MNIST_DC_GAN.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "MNIST_DC_GAN",
      "version": "0.3.2",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "ShnVJdsVPWvr",

This file has been truncated. show original

Hope this resolved your doubt.
Plz mark the doubt as resolved in my doubts section.

abhishek13yadv · April 16, 2020, 3:26pm

It shows an error
rewatch it.

FailedPreconditionError: Error while reading resource variable _AnonymousVar41 from Container: localhost. This could mean that the variable was uninitialized. Not found: Resource localhost/_AnonymousVar41/N10tensorflow3VarE does not exist. [[node mul_18/ReadVariableOp (defined at /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3009) ]] [Op:__inference_keras_scratch_graph_1772]Function call stack:keras_scratch_graph site:stackoverflow.com

S18CRX0120 · April 16, 2020, 5:43pm

Hey @abhishek13yadv, actually now colab provides tensorflow 2.0, so keras functions poses a problem in it,

https://colab.research.google.com/drive/1A_1l3h2kMQ13EGuILC8NhQi6xA5ZM4wm

Use this code for reference. Major changes are in import statements, and swtiching off the default eager execution in tensorflow.

Hope this resolved your doubt.
Plz mark the doubt as resolved in my doubts section.

S18CRX0120 · April 20, 2020, 6:19pm

I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.

On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.