All In One: AttributeError: module 'a' has no attribute 'b'

This tutorial, as its name, will show you common AttributeError: module has no attribute errors and how to fix them.

You can use Ctrl + F to search for the error you are facing.

For your reference:

AttributeError: module 'keras.backend' has no attribute 'set_session'

• Solution: Change keras.backend.set_session to tf.compat.v1.keras.backend.set_session.

• Example:

  before

  import tensorflow as tf
  import keras.backend as K
  

  after

  import tensorflow as tf
  import tensorflow.compat.v1.keras.backend as K
  

AttributeError: module 'tensorflow' has no attribute 'Session'

• Solution: Change tf.Session to tf.compat.v1.Session.

• Example:

  before

  import tensorflow as tf
  sess = tf.Session()
  

  after

  import tensorflow as tf
  sess = tf.compat.v1.Session()
  

AttributeError: module 'tensorflow' has no attribute 'placeholder'

• Solution: Change tf.placeholder to tf.compat.v1.placeholder. Sometimes will need to add tf.compat.v1.disable_eager_execution().

• Example:

  before

  import tensorflow as tf
  x = tf.placeholder(tf.float32, shape=(None, 784))
  

  after

  import tensorflow as tf
  tf.compat.v1.disable_eager_execution()
  x = tf.compat.v1.placeholder(tf.float32, shape=(None, 784))
  

AttributeError: module 'tensorflow' has no attribute 'get_default_graph'

• Solution: Change tf.get_default_graph to tf.compat.v1.get_default_graph.

• Example:

  before

  import tensorflow as tf
  graph = tf.get_default_graph()
  

  after

  import tensorflow as tf
  graph = tf.compat.v1.get_default_graph()
  

AttributeError: module 'torch' has no attribute '_six'

• Solution: Update torch to the latest version and restart the notebook kernel.
• To update torch: run pip install --upgrade torch torchvision.

Conclusion

Thank you for reading. I will try to put a search bar so that you can search through the errors better.