module 'tensorflow.python.keras.datasets.fashion_mnist' has no attribute 'load_data'

Question

I am currently following this intro tutorial on the Keras website: https://www.tensorflow.org/tutorials/keras/basic_classification

Several steps in I run into this error after calling fashion_mnist.load_data():

AttributeError: module 'tensorflow.python.keras.datasets.fashion_mnist' has no attribute 'load_data'

This is the full output:

Python 3.6.6 (v3.6.6:4cf1f54eb7, Jun 27 2018, 03:37:03) [MSC v.1900 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
>>> from tensorflow import keras
>>> fashion_mnist = keras.datasets.fashion_mnist
>>> (train_images, train_labels), (test_images, test_labels) = fashion_mnist.load_data()
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
AttributeError: module 'tensorflow.python.keras.datasets.fashion_mnist' has no attribute 'load_data'

I'm using tensorflow 1.5.0, Keras 2.2.2, and Python 3.6.6.

Is tensorflow's tutorial outdated, or am I missing something? If I use the mnist set instead of fashion_mnist, it works with no problem. From this link https://www.tensorflow.org/api_docs/python/tf/keras/datasets/fashion_mnist it would seem that fashion_mnist does indeed have a function called load_data.

Answer 1

The problem lies indeed in your Tensorflow version. The tutorial you link to uses version 1.9.0:

print(tf.__version__)
# 1.9.0

which does include a function load_data for fashion_mnist (docs). But this function is missing from your version, as you can see from the v1.5 docs.