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

[JuliaH22]

---

AttributeError Traceback (most recent call last)
in
1 get_ipython().run_line_magic('matplotlib', 'inline')
2 import matplotlib.pyplot as plt
----> 3 import tensorflow as tf
4 import numpy as np
5 import pandas as pd

~\miniconda3\envs\tensorflow\lib\site-packages\tensorflow_init_.py in
99
100 # We still need all the names that are toplevel on tensorflow_core
--> 101 from tensorflow_core import *
102
103 # These should not be visible in the main tf module.

~\miniconda3\envs\tensorflow\lib\site-packages\tensorflow_core_init_.py in
44 from . _api.v2 import autograph
45 from . _api.v2 import bitwise
---> 46 from . _api.v2 import compat
47 from . _api.v2 import config
48 from . _api.v2 import data

~\miniconda3\envs\tensorflow\lib\site-packages\tensorflow_core_api\v2\compat_init_.py in
37 import sys as _sys
38
---> 39 from . import v1
40 from . import v2
41 from tensorflow.python.compat.compat import forward_compatibility_horizon

~\miniconda3\envs\tensorflow\lib\site-packages\tensorflow_core_api\v2\compat\v1_init_.py in
30 from . import autograph
31 from . import bitwise
---> 32 from . import compat
33 from . import config
34 from . import data

~\miniconda3\envs\tensorflow\lib\site-packages\tensorflow_core_api\v2\compat\v1\compat_init_.py in
37 import sys as _sys
38
---> 39 from . import v1
40 from . import v2
41 from tensorflow.python.compat.compat import forward_compatibility_horizon

~\miniconda3\envs\tensorflow\lib\site-packages\tensorflow_core_api\v2\compat\v1\compat\v1_init_.py in
27
28 from . import compat
---> 29 from tensorflow._api.v2.compat.v1 import app
30 from tensorflow._api.v2.compat.v1 import audio
31 from tensorflow._api.v2.compat.v1 import autograph

~\miniconda3\envs\tensorflow\lib\site-packages\tensorflow_core_api\v2\compat_init_.py in
37 import sys as _sys
38
---> 39 from . import v1
40 from . import v2
41 from tensorflow.python.compat.compat import forward_compatibility_horizon

~\miniconda3\envs\tensorflow\lib\site-packages\tensorflow_core_api\v2\compat\v1_init_.py in
30 from . import autograph
31 from . import bitwise
---> 32 from . import compat
33 from . import config
34 from . import data

~\miniconda3\envs\tensorflow\lib\site-packages\tensorflow_core_api\v2\compat\v1\compat_init_.py in
37 import sys as _sys
38
---> 39 from . import v1
40 from . import v2
41 from tensorflow.python.compat.compat import forward_compatibility_horizon

~\miniconda3\envs\tensorflow\lib\site-packages\tensorflow_core_api\v2\compat\v1\compat\v1_init_.py in
665 _current_module = _sys.modules[name]
666 try:
--> 667 from tensorflow_estimator.python.estimator.api._v1 import estimator
668 _current_module.path = (
669 [_module_util.get_parent_dir(estimator)] + _current_module.path)

~\AppData\Roaming\Python\Python37\site-packages\tensorflow_estimator_init_.py in
8 import sys as _sys
9
---> 10 from tensorflow_estimator._api.v1 import estimator
11
12 del _print_function

~\AppData\Roaming\Python\Python37\site-packages\tensorflow_estimator_api\v1\estimator_init_.py in
8 import sys as _sys
9
---> 10 from tensorflow_estimator._api.v1.estimator import experimental
11 from tensorflow_estimator._api.v1.estimator import export
12 from tensorflow_estimator._api.v1.estimator import inputs

~\AppData\Roaming\Python\Python37\site-packages\tensorflow_estimator_api\v1\estimator\experimental_init_.py in
8 import sys as _sys
9
---> 10 from tensorflow_estimator.python.estimator.canned.dnn import dnn_logit_fn_builder
11 from tensorflow_estimator.python.estimator.canned.kmeans import KMeansClustering as KMeans
12 from tensorflow_estimator.python.estimator.canned.linear import LinearSDCA

~\AppData\Roaming\Python\Python37\site-packages\tensorflow_estimator\python\estimator\canned\dnn.py in
31 from tensorflow.python.keras.utils import losses_utils
32 from tensorflow.python.util.tf_export import estimator_export
---> 33 from tensorflow_estimator.python.estimator import estimator
34 from tensorflow_estimator.python.estimator.canned import head as head_lib
35 from tensorflow_estimator.python.estimator.canned import optimizers

~\AppData\Roaming\Python\Python37\site-packages\tensorflow_estimator\python\estimator\estimator.py in
51 from tensorflow_estimator.python.estimator import model_fn as model_fn_lib
52 from tensorflow_estimator.python.estimator import run_config
---> 53 from tensorflow_estimator.python.estimator import util as estimator_util
54 from tensorflow_estimator.python.estimator.export import export_lib
55 from tensorflow_estimator.python.estimator.mode_keys import ModeKeys

~\AppData\Roaming\Python\Python37\site-packages\tensorflow_estimator\python\estimator\util.py in
73
74
---> 75 class _DatasetInitializerHook(tf.compat.v1.train.SessionRunHook):
76 """Creates a SessionRunHook that initializes the passed iterator."""
77

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

[JuliaH22]

Hi everyone
after a long attempt to install tensorflow gpu version by following this tutorial:
https://www.youtube.com/watch?v=qrkEYf-YDyI&t=295s
I always find myself facing the above error. Can someone please help me?

[ymodak]

What is the TF version you are trying to install? Also what steps did you follow?

[AnimeshMaheshwari22]

You can try uninstalling and re-installing TensorFlow. Also, check this link https://www.tensorflow.org/guide/migrate

[JuliaH22]

What is the TF version you are trying to install? Also what steps did you follow?

At first I uninstalled anaconda and all the NVIDIA programs I had before, to start over from scratch.
Then I followed every step of this tutorial in detail:
https://www.youtube.com/watch?v=qrkEYf-YDyI
Step 1: I have NVIDIA Geforce GTX 1050 Ti, I installed the GTX 1650 driver
Step 2: I downloaded the updated Setup Script provided in https://github.com/jeffheaton/t81_558_deep_learning
To search for compatible version of Tensorflow 2.1.0: https://www.tensorflow.org/install/gpu?hl=fr
Step 3: I installed CUDA Toolkit 10.1
Step 4: I installed cuDNN 7.6.5 for Cuda 10.1
I've added all the necessary paths in my path environment
Step 5: I installed TensorRT 6.0
Step 6: Once I installed miniconda, i did the following setup commands using my command prompt:
-conda
-conda install jupyter
-Once jupyter installation is done, I run the installaton script downloaded in step 2 using this command dir *.yml.
Then, i run this command: conda env create -v -f tensorflow-gpu.yml, It took me a long time.
Later I used "conda env create -f tensorflow-gpu.yml". Again, several things have been installed.
Just after i run: python -m ipykernel install --user --name tensorflow --display-name "Python 3.7 (tensorflow)" . Then python
Once tensorflow activated using "conda activate tensorflow", I get the error on top when I try to import Tensorflow as tf

[JuliaH22]

You can try uninstalling and re-installing TensorFlow. Also, check this link https://www.tensorflow.org/guide/migrate

I'm still getting the same error, even with the reinstallation of tensorflow using pip.

[AnimeshMaheshwari22]

You can follow the steps mentioned in the article below and check if it works.
https://towardsdatascience.com/tensorflow-gpu-installation-made-easy-use-conda-instead-of-pip-52e5249374bc

If it does not work, try pip install tensorflow==2.1. Found this here: #37525.

You can try uninstalling and re-installing TensorFlow. Also, check this link https://www.tensorflow.org/guide/migrate

I'm still getting the same error, even with the reinstallation of tensorflow using pip.

[JuliaH22]

You can follow the steps mentioned in the article below and check if it works.
https://towardsdatascience.com/tensorflow-gpu-installation-made-easy-use-conda-instead-of-pip-52e5249374bc

If it does not work, try pip install tensorflow==2.1. Found this here: #37525.

You can try uninstalling and re-installing TensorFlow. Also, check this link https://www.tensorflow.org/guide/migrate

I'm still getting the same error, even with the reinstallation of tensorflow using pip.

Same problem even with pip install tensorflow==2.1 ! Sorry, i already saw that issue

[JuliaH22]

the steps mentioned in the article

Thank you for this article. I will follow these steps

[mihaimaruseac]

Please properly fill in issue template and use the proper Markdown formatting around errors and code blocks.

[google-ml-butler]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you.

[ltorres321]

I have the same exact issue Here is my post I cant find a working solution for this issue
https://stackoverflow.com/questions/62622704/attributeerror-module-tensorflow-has-no-attribute-compat-when-loading-tf-co
In google I see post to install MV C++ i already have VS 2019 installed

[mindUpGit]

Same issue for us.
We have installed the same combo on another machine ,,, it works.
Here we have a notebook with internal graphics card and one additional GTX 1650
GTX driver: 442.70
CUDA 10.1
cuDNN 7.6.5.32
TensorRT 6.0
system path changed like specified
We are using anaconda for test purpose.
packages installed all via conda install - no pip.

[mindUpGit]

ISSUE FIXED for me

Indeed the command "conda install tensorflow-gpu==2.1.0" installed version 2.2.0 of tensorflow-estimator.
After "conda install tensorflow-estimator==2.1.0" everything works fine

[amahendrakar]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you.

Automatically closing due to lack of recent activity. Please update the issue when new information becomes available, and we will reopen the issue. Thanks!

[google-ml-butler]

Are you satisfied with the resolution of your issue?
Yes
No

[mikechen66]

After "conda install tensorflow-estimator==2.1.0", everything works fine for me too .