AttributeError: module tensorflow has no attribute contrib #7767

Girish1127 · 2019-11-06T17:24:05Z

version: python 3.6.9
OS: windows10
IDE: anaconda, spyder (executing locally)
also installed tensorflow by creating new environment in anaconda (without using pip)
tensorflow version 2.0.0

code shown below:
import tensorflow as tf
''''
'''''
init_kernel = tf.contrib.layers.xavier_initializer()
getting error for the above line.
AttributeError: module 'tensorflow' has no attribute 'contrib'

gaseosaluz · 2019-11-06T21:23:38Z

Since you say that you used TF 2.0, from the TF docs:

https://www.tensorflow.org/guide/upgrade#compatibility_modules

I gather (but I could be wrong) that tf.contrib is not available on TF 2.0 (At least not all of the functionality).

These discussion maybe helpful?
tensorflow/tensorflow#26854
tensorflow/community#18

sinakhani · 2019-12-05T20:29:51Z

Hello there,
When I use
initializer = tf.contrib.layers.xavier_initializer(seed = 0))
to initialize my ML layer I get the following error
AttributeError: module 'tensorflow' has no attribute 'contrib'
Any suggestion to remove this error is really appreciated.

perturbedmuffin · 2020-01-16T21:29:25Z

tensorflow.contrib was removed in TF2. Tensorflow Core released a utility to convert your TF 1.x scripts to TF2.0 which can be found here: https://www.tensorflow.org/guide/upgrade

EDIT: It was brought to my attention that the utility mentioned here cannot convert the tf.contrib library, so this method does not work. I will leave the comment up to save someone's time in the future.

mrhanti · 2020-02-24T17:28:04Z

@EduardoABarrera parts of tensorflow/models latest code base are still using the TF 1.x API.

For instance:

...
from object_detection.utils import static_shape

// Throws: AttributeError: module 'tensorflow' has no attribute 'contrib'
slim = tf.contrib.slim

BOX_ENCODINGS = box_predictor.BOX_ENCODINGS
...

Find the complete file here

This is causing problems when doing transfer learning on object detection models. Do you have any idea or workaround for this one?

perturbedmuffin · 2020-02-24T19:50:10Z

@EduardoABarrera parts of tensorflow/models latest code base are still using the TF 1.x API.

For instance:
...
from object_detection.utils import static_shape

// Throws: AttributeError: module 'tensorflow' has no attribute 'contrib'
slim = tf.contrib.slim

BOX_ENCODINGS = box_predictor.BOX_ENCODINGS
...
Find the complete file here

This is causing problems when doing transfer learning on object detection models. Do you have any idea or workaround for this one?

@mrhanti I also bumped into this problem, and it turns out that the library tf.contrib is the only library that isn't converted via the utility mentioned above. If you're trying to convert a model like AlexNet that was written in TF1.X to a version of AlexNet written in TF2.0, I don't think there is a simple solution for that. I detailed this same problem (almost?) in this Stack overflow post.

The third point in my Stack overflow post is what I had to end up doing (writing out the model architecture by hand after converting it a few times), and it involved a lot of work. Sorry to bear bad news. :(

init-22 · 2020-04-24T08:16:25Z

So there is no easy way out for this? @EduardoABarrera

pbanikk · 2020-08-01T15:48:46Z

https://www.tensorflow.org/api_docs/python/tf/compat/v1/nn/rnn_cell/BasicLSTMCell
The only workaround if you want to use Basic LSTM Cell.

pourya-farzi · 2020-08-19T09:30:14Z

The unwinding path is just this line :
!pip install tensorflow==1.14
better than getting involved in convoluted procedure

mikechen66 · 2020-10-20T02:56:40Z

For the usage in LSTM, it has been changed to the following.

import tensorflow.compat.v1 as tf
tf.disable_v2_behavior() 
# -outputs, states = tf.contrib.rnn.static_rnn(lstm_cells, _X, dtype=tf.float32)
outputs, states = tf.compat.v1.nn.static_rnn(lstm_cells, _X, dtype=tf.float32)

And then it works. However, contrib is used widely in TensorFlow 1.x, Google does no provides an integrated solution to the contrib issue. For instance, it is another solution to the specific contrib case.

# -initializer = tf.contrib.layers.xavier_initializer(seed = 1)
initializer = tf.truncated_normal_initializer(stddev=0.1)

It is the pain of TensorFlow 2.x by the Google Team. Therefore, we need to solve the contrib problems case by case.

saookami · 2020-10-20T08:41:42Z

My Tensorflow version is 2.1.0，I checked file where in tensorflow\contrib.
There's just a little bit of code here.
Like i found slim film but there's have nothion in it,

pnmartinez · 2021-01-15T17:28:11Z

The unwinding path is just this line :
!pip install tensorflow==1.14
better than getting involved in convoluted procedure

I would add: create a specific environment for this and use it for TF1 stuff. You can do that in conda like:

conda create --name tf1 python=3.7  # python 3.8 won't work!

Then you can happily:

conda install tensorflow==1.14

ShaikhaTheGreen · 2021-02-17T03:48:11Z

I have a similar problem with this line:

        self.optimizer = tf.compat.v1.contrib.opt.ScipyOptimizerInterface(self.loss, 
                                                                method = 'L-BFGS-B', 
                                                                options = {'maxiter': 50000,
                                                                           'maxfun': 50000,
                                                                           'maxcor': 50,
                                                                           'maxls': 50,
                                                                           'ftol' : 1.0 * np.finfo(float).eps})
        
        init = tf.global_variables_initializer()
        self.sess.run(init)

It throws:
AttributeError: module 'tensorflow._api.v2.compat.v1' has no attribute 'contrib'

Any suggestion is appreciated. Thanks!

gholn · 2021-02-17T12:14:30Z

Please check if your PYTHONPATH is set. If it is not set, you can get this error of 'has no attribute 'contrib''.
In Windows, set PYTHONPATH in System Environment variables.
In Linus, set it by $export PYTHONPATH=$PYTHONPATH:pwd:pwd/slim
Or you can add it to '.bashrc' file

sandeep22-v · 2021-03-10T04:38:09Z

export PYTHONPATH=$PYTHONPATH:pwd:pwd/slim
This will help surely..thank me later

Fado2021 · 2021-03-16T10:48:52Z

@engsbk Hello,
Please how do you solved your issue??

AttributeError: module 'tensorflow._api.v2.compat.v1' has no attribute 'contrib'

ShaikhaTheGreen · 2021-03-16T12:00:53Z

@engsbk Hello,
Please how do you solved your issue??

AttributeError: module 'tensorflow._api.v2.compat.v1' has no attribute 'contrib'

You can solve this in one of two ways:

you can entirely remove Tensorflow v1.x and install Tensorflow v2.x
Or Install Tensorflow v2.x and make sure it in the PATH directory.

Unless the code you are trying to run is totally uncompatible with Tensorflow v.2.x, in which case just switch back to Tensorflow v1.x.

Fado2021 · 2021-03-16T13:59:39Z

Thanks you. Le mar. 16 mars 2021 à 20:01, Shaikha Alkhuder ***@***.***> a écrit :

…

@engsbk <https://github.com/engsbk> Hello, Please how do you solved your issue?? AttributeError: module 'tensorflow._api.v2.compat.v1' has no attribute 'contrib' You can solve this in one of two ways: - you can entirely remove Tensorflow v1.x and install Tensorflow v2.x - Or Install Tensorflow v2.x and make sure it in the PATH directory. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#7767 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATHS5ZRTTFXCQ2HYUG4VDJ3TD5CAVANCNFSM4JJYRFUA> .

Fado2021 · 2021-03-17T05:03:39Z

@engsbk I tried your suggestion, but still have problems. Any help??

TensorFlow 2.0.0 + Keras 2.3.1 on Python 3.6.

Problem1: config = tf.ConfigProto()
AttributeError: module 'tensorflow' has no attribute 'ConfigProto'

Problem2: self.optimizer = tf.contrib.opt.ScipyOptimizerInterface(self.loss,
method='L-BFGS-B',
options={'maxiter': 20000,
'maxfun': 50000,
'maxcor': 50,
'maxls': 50,
'ftol': 1 * np.finfo(float).eps})
AttributeError: module 'tensorflow_core.compat.v1' has no attribute 'contrib'

fdsig · 2021-06-25T18:52:17Z

On Linux only:

`sudo apt update

update-alternatives --install /usr/bin/python python /usr/opt/python3.7  

virtualenv -p python3.7 contrib/**insert your path/and name for virtual environment)

source **insert/path/to/environment**/bin/activate 

pip install tensorflow==1.** `

As mentioned above this wont work on python 3.7 or later.

useful links are:

https://www.tensorflow.org/guide/migrate

Python alternative install on Linux:

https://medium.com/analytics-vidhya/how-to-install-and-switch-between-different-python-versions-in-ubuntu-16-04-dc1726796b9b

https://hackersandslackers.com/multiple-versions-python-ubuntu/

python source download urls:

https://www.python.org/ftp/python/3.7.0/

managing virtual environments Linux:

https://www.liquidweb.com/kb/creating-virtual-environment-ubuntu-16-04/

https://docs.python.org/3/library/venv.html

haveing tried most of the above- including migrating code and import tensorflow.compat.v1 as tf tf.disable_v2_behavior() which both are not viable workaround for code that uses the config module.

I was sucesfull able to run code with the above method without issue.

Hope this helps :-)

Leci37 · 2023-01-09T01:03:38Z

Replace TF V1 line
slim = tf.contrib.slim
for TF V2
import tf_slim as slim

haroon445 · 2023-08-05T06:05:42Z

version: python 3.6.9 OS: windows10 IDE: anaconda, spyder (executing locally) also installed tensorflow by creating new environment in anaconda (without using pip) tensorflow version 2.0.0

code shown below: import tensorflow as tf '''' ''''' init_kernel = tf.contrib.layers.xavier_initializer() getting error for the above line. AttributeError: module 'tensorflow' has no attribute 'contrib'

The "attributeerror: module tensorflow has no attribute contrib" error occurs in Python when we try to import or use tf.contrib in TensorFlow 2.0 or later versions. This is because tf.contrib has been removed from TensorFlow 2.0.

To fix this error we need to downgrade the tensorflow to the version that supports the tf.contrib or upgrade the code from 1.x version to 2.x version. You can check this guide to resolve this error, I have found it very helpful.

amkjitu · 2023-08-27T23:08:20Z

Replace TF V1 line slim = tf.contrib.slim for TF V2 import tf_slim as slim

THANKSSSSS IT WORKSSSS

shravani1115 · 2024-04-03T09:30:07Z

Replace TF V1 line slim = tf.contrib.slim for TF V2 import tf_slim as slim

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

I am facing this error please help

Roxie2003 · 2024-04-15T16:48:24Z

Replace TF V1 line slim = tf.contrib.slim for TF V2 import tf_slim as slim

Thank you so muchhhhh :)

RubensFPereira · 2024-04-18T06:37:53Z

Hello Everyone!
You guys can help me with this problem?

File "F:\Anaconda\envs\pothole\lib\site-packages\object_detection-0.1-py3.6.egg\object_detection\data_decoders\tf_example_decoder.py", line 332, in
class _BackupHandler(tf_example_decoder.ItemHandler):
AttributeError: module 'object_detection.data_decoders.tf_example_decoder' has no attribute 'ItemHandler'

Thanks!

saberkun closed this as completed Nov 6, 2019

HACLANG mentioned this issue Dec 4, 2021

ModuleNotFoundError: No module named 'gym.spaces.dict_space' openai/multi-agent-emergence-environments#20

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

AttributeError: module tensorflow has no attribute contrib #7767

AttributeError: module tensorflow has no attribute contrib #7767

Girish1127 commented Nov 6, 2019

gaseosaluz commented Nov 6, 2019

sinakhani commented Dec 5, 2019

perturbedmuffin commented Jan 16, 2020 •

edited

Loading

mrhanti commented Feb 24, 2020

perturbedmuffin commented Feb 24, 2020

init-22 commented Apr 24, 2020

pbanikk commented Aug 1, 2020

pourya-farzi commented Aug 19, 2020

mikechen66 commented Oct 20, 2020 •

edited

Loading

saookami commented Oct 20, 2020

pnmartinez commented Jan 15, 2021 •

edited

Loading

ShaikhaTheGreen commented Feb 17, 2021

gholn commented Feb 17, 2021

sandeep22-v commented Mar 10, 2021

Fado2021 commented Mar 16, 2021

ShaikhaTheGreen commented Mar 16, 2021 •

edited

Loading

Fado2021 commented Mar 16, 2021 via email

Fado2021 commented Mar 17, 2021

fdsig commented Jun 25, 2021 •

edited

Loading

Leci37 commented Jan 9, 2023

haroon445 commented Aug 5, 2023

amkjitu commented Aug 27, 2023

shravani1115 commented Apr 3, 2024

Roxie2003 commented Apr 15, 2024

RubensFPereira commented Apr 18, 2024 •

edited

Loading

AttributeError: module tensorflow has no attribute contrib #7767

AttributeError: module tensorflow has no attribute contrib #7767

Comments

Girish1127 commented Nov 6, 2019

gaseosaluz commented Nov 6, 2019

sinakhani commented Dec 5, 2019

perturbedmuffin commented Jan 16, 2020 • edited Loading

mrhanti commented Feb 24, 2020

perturbedmuffin commented Feb 24, 2020

init-22 commented Apr 24, 2020

pbanikk commented Aug 1, 2020

pourya-farzi commented Aug 19, 2020

mikechen66 commented Oct 20, 2020 • edited Loading

saookami commented Oct 20, 2020

pnmartinez commented Jan 15, 2021 • edited Loading

ShaikhaTheGreen commented Feb 17, 2021

gholn commented Feb 17, 2021

sandeep22-v commented Mar 10, 2021

Fado2021 commented Mar 16, 2021

ShaikhaTheGreen commented Mar 16, 2021 • edited Loading

Fado2021 commented Mar 16, 2021 via email

Fado2021 commented Mar 17, 2021

fdsig commented Jun 25, 2021 • edited Loading

Leci37 commented Jan 9, 2023

haroon445 commented Aug 5, 2023

amkjitu commented Aug 27, 2023

shravani1115 commented Apr 3, 2024

Roxie2003 commented Apr 15, 2024

RubensFPereira commented Apr 18, 2024 • edited Loading

perturbedmuffin commented Jan 16, 2020 •

edited

Loading

mikechen66 commented Oct 20, 2020 •

edited

Loading

pnmartinez commented Jan 15, 2021 •

edited

Loading

ShaikhaTheGreen commented Mar 16, 2021 •

edited

Loading

fdsig commented Jun 25, 2021 •

edited

Loading

RubensFPereira commented Apr 18, 2024 •

edited

Loading