Module 'tensorflow' has no attribute 'contrib'

I am trying to train my own custom object detector using Tensorflow Object-Detection-API

I installed the tensorflow using "pip install tensorflow" in my google compute engine. Then I followed all the instructions on this site: https://tensorflow-object-detection-api-tutorial.readthedocs.io/en/latest/training.html

When I try to use train.py I am getting this error message:

>Traceback (most recent call last): > File "train.py", line 49, in > from object_detection.builders import dataset_builder >File "/usr/local/lib/python3.6/dist-packages/object_detection-0.1->py3.6.egg/object_detection/builders/dataset_builder.py", line 27, in >from object_detection.data_decoders import tf_example_decoder >File "/usr/local/lib/python3.6/dist-packages/object_detection-0.1-py3.6.egg/object_detection/data_decoders/tf_example_decoder.py", line 27, in slim_example_decoder = tf.contrib.slim.tfexample_decoder AttributeError: module 'tensorflow' has no attribute 'contrib'

Also I am getting different results when I try to learn version of tensorflow.

>python3 -c 'import tensorflow as tf; print(tf.version)' : 2.0.0-dev20190422

and when I use

>pip3 show tensorflow:

>Name: tensorflow Version: 1.13.1 Summary: TensorFlow is an open source machine learning framework for everyone. Home-page: https://www.tensorflow.org/ Author: Google Inc. Author-email: [email protected] License: Apache 2.0 Location: /usr/local/lib/python3.6/dist-packages Requires: gast, astor, absl-py, tensorflow-estimator, keras-preprocessing, grpcio, six, keras-applications, wheel, numpy, tensorboard, protobuf, termcolor Required-by:

    sudo python3 train.py --logtostderr --train_dir=training/ -- 
    pipeline_config_path=training/ssd_inception_v2_coco.config

What should I do to solve this problem? I couldn't find anything about this error message except this: https://stackoverflow.com/questions/38238192/tensorflow-module-object-has-no-attribute-contrib

tf.contrib has moved out of TF starting TF 2.0 alpha.
Take a look at these tf 2.0 release notes https://github.com/tensorflow/tensorflow/releases/tag/v2.0.0-alpha0
You can upgrade your TF 1.x code to TF 2.x using the tf_upgrade_v2 script https://www.tensorflow.org/alpha/guide/upgrade

This issue might be helpful for you, it explains how to achieve TPUStrategy, a popular functionality of tf.contrib in TF<2.0.

So, in TF 1.X you could do the following:

resolver = tf.contrib.cluster_resolver.TPUClusterResolver('grpc://' + os.environ['COLAB_TPU_ADDR'])
tf.contrib.distribute.initialize_tpu_system(resolver)
strategy = tf.contrib.distribute.TPUStrategy(resolver)

And in TF>2.0, where tf.contrib is deprecated, you achieve the same by:

tf.config.experimental_connect_to_host('grpc://' + os.environ['COLAB_TPU_ADDR'])
resolver = tf.distribute.cluster_resolver.TPUClusterResolver('grpc://' + os.environ['COLAB_TPU_ADDR'])
tf.tpu.experimental.initialize_tpu_system(resolver)
strategy = tf.distribute.experimental.TPUStrategy(resolver) 

I used google colab to run my models and everything was perfect untill i used inline tesorboard. With tensorboard inline, I had the same issue of "Module 'tensorflow' has no attribute 'contrib'".

It was able to run training when rebuild and reinstall the model using setup.py(research folder) after initialising tensorboard.

I used tensorflow 1.8 to train my model and there is no problem for now. Tensorflow 2.0 alpha is not suitable with object detection API

I'm using Google Colab as well. A comment suggested to put

%tensorflow_version 1.x

in the first (code) cell, and it worked!

One easy way is you can pass your code written in TensorFlow 1.x to the below code to automatically upgrade it to TensorFlow 2.x.

$tf_upgrade_v2 \
--intree my_project/ \
--outtree my_project_v2/ \
--reportfile report.txt

The above code will replace all the commands which are deprecated in 2.x with the onces that are actually working in 2.x. And then you can run your code in TensorFlow 2.x.

In case if it throws an error and is unable to convert the complete code and then don't panic. Please open the "report.txt" file that is generated by the above code. In this file, you will find commands that are deprecated and their alternative commands that can be used in TensorFlow 2.x.

Taadaa, just replace the commands that are throwing errors with the new ones.

Example:

If the command in TensorFlow 1.x is:

tf.contrib

Then the same command in Tensorflow 2.x is:

tf.compat.v1.estimator

In the above example replace "tf.contrib" with "tf.compat.v1.estimator" and that should solve the problem.

If you want to use tf.contrib, you need to now copy and paste the source code from github into your script/notebook. It's annoying and doesn't always work. But that's the only workaround I've found. For example, if you wanted to use tf.contrib.opt.AdamWOptimizer, you have to copy and paste from here. https://github.com/tensorflow/tensorflow/blob/590d6eef7e91a6a7392c8ffffb7b58f2e0c8bc6b/tensorflow/contrib/opt/python/training/weight_decay_optimizers.py#L32

I face the same error and solve it by install python version 3.7 then i can install tensorflow 1.15 and it work.

For me it worked using the latest release of tensorflow: pip install tensorflow==2.2.0