Select Page

How to Solve Python AttributeError: module ‘tensorflow’ has no attribute ‘placeholder’

by | Programming, Python, Tips

In TensorFlow 2.0, tf.placeholder is no longer in use. A placeholder is a variable that we will assign data to at a later point. It allows us to create operations and build the computation graph without the data.

In TensorFlow 2.0, we can use tf.function to execute graph operations eagerly.

If you want to continue using placeholder in TensorFlow 2.0, use tf.compat.v1.placeholder() instead.

You can follow the migration guide to migrate your TensorFlow code from TensorFlow 1.x to TensorFlow 2.

This tutorial will go through the error in detail and how to solve it with code examples.

AttributeError: module ‘tensorflow’ has no attribute ‘placeholder’

AttributeError occurs in a Python program when we try to access an attribute (method or property) that does not exist for a particular object. The part “‘module ‘tensorflow’ has no attribute ‘placeholder’” tells us that the TensorFlow module does not have the attribute placeholder(). The placeholder() function belongs to the TensorFlow 1.x API.

Generally, if the AttributeError refers to a module not having an attribute, either the functionality is under a different name or deprecated. Consult the documentation of the module to find where functionalities and sub-modules are.

Do not name python scripts after module names. For example, naming a script If you try:

import tensorflow as tf

you will import the script file under your current working directory, rather than the actual TensorFlow module. The Python interpreter searches for a module first in the current working directory, then the PYTHONPATH, then the installation-dependent default path. You can name a script after its functionality instead.

What is a TensorFlow Placeholder?

A Tensorflow placeholder is a variable that holds the place of data that we assign at a later point. Using placeholders allows us to create the computation graph and operations without the requirement of data. Once we create a session, we feed the data into the placeholder. Let’s look at the syntax of the placeholder:

tf.compat.v1.placeholder(dtype, shape=None, name=None)


  • dtype: Required. The data type of the elements to feed into the tensor
  • shape: Optional. Tensor shape. Default is None.
  • name: Optional. Name of the operation. Default is None.


  • A Tensor to feed a value into


Let’s look at an example where we try to use a placeholder in TensorFlow 2.0:

# importing Tensorflow
import tensorflow as tf


# Define a placeholder
a = tf.placeholder(tf.float32, None)
# Define an operation
b = a + 10
# Session as context manager
with tf.Session() as session:
    # Feed data to placeholder

    operation_res =, feed_dict={a: [10, 20, 30, 40]})

    print("result: " + str(operation_res))

In the above code, the placeholder has a dtype of tf.float32 and set it not to have a specific size.

We create the operation before feeding in the data. The operation adds 10 to the tensor.

We will feed the values into the TensorFlow placeholder using feed_dict when calling Let’s run the code to see the result:

AttributeError                            Traceback (most recent call last)
<ipython-input-1-0c4d9505b527> in <module>
      6 # Define a placeholder
----> 7 a = tf.placeholder(tf.float32, None)
      9 # Define an operation

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

The error occurs because the placeholder function is no longer present in TensorFlow 2.0.

Solution #1: Use tf.compat.v1

We can use the tf.compat.v1 module to solve this error. The module contains the complete TF1.x API with its original semantics. Generally, you should avoid using the legacy compat.v1 APIs for any new code you write in TensorFlow 2.0, but this approach is suitable for previously written code. Let’s look at the revised code:

# importing Tensorflow
import tensorflow.compat.v1 as tf
# disabling eager mode

# Define a placeholder
a = tf.placeholder(tf.float32, None)
# Define an operation
b = a + 10
# Session as context manager
with tf.Session() as session:
    # Feed data to placeholder
    operation_res =, feed_dict={a: [10, 20, 30, 40]})
    print("result: " + str(operation_res))

We also need to disable Eager mode, which is the default in TensorFlow 2.0. We want to execute the operations using the TensorFlow 1.x Session-based paradigm. Let’s run the code to see the result:

result: [20. 30. 40. 50.]

Solution #2: Use tf.function

TensorFlow 2 uses functions instead of sessions, which integrates better with Python runtime. tf.function compiles a function into a callable TensorFlow graph. We can define a function with the decorator @tf.function or we can make a direct call using tf.function.

The input parameters of the function take the place of placeholders.

Let’s look at how to add two numbers using tf.function.

import tensorflow as tf
def add_func(x):
    y = x + 10
    return y

x = tf.constant([10, 20, 30, 40])

result = add_func(x)


Let’s run the code to see the result:

tf.Tensor([20 30 40 50], shape=(4,), dtype=int32)

We successfully used the function to add 10 to every element in the tensor.

TensorFlow 1.x vs TensorFlow 2

TensorFlow 2 follows a fundamentally different programming paradigm from TensorFlow 1.x. There are different runtime behaviors around execution, variables, control flow, tensor shapes, and tensor equality comparisons. TensorFlow 2 is preferable to use as it removes redundant APIs and makes APIs more consistent.

To migrate to TensorFlow 2, follow the TF1.x to TF2 migration guide.


Congratulations on reading to the end of this tutorial! The AttributeError: module ‘tensorflow’ has no attribute ‘placeholder’ occurs when you try to use a placeholder in TensorFlow 2.0. placeholder is part of the TF1.x API. To solve this error, you can either migrate to TensorFlow 2.0 and use tf.function or use tf.compat.v1.Session().

For further reading on TensorFlow, go to the articles:

Go to the online courses page on Python to learn more about Python for data science and machine learning.

Have fun and happy researching!

Research Scientist at Moogsoft | + posts

Suf is a research scientist at Moogsoft, specializing in Natural Language Processing and Complex Networks. Previously he was a Postdoctoral Research Fellow in Data Science working on adaptations of cutting-edge physics analysis techniques to data-intensive problems in industry. In another life, he was an experimental particle physicist working on the ATLAS Experiment of the Large Hadron Collider. His passion is to share his experience as an academic moving into industry while continuing to pursue research. Find out more about the creator of the Research Scientist Pod here and sign up to the mailing list here!