Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

AttributeError: module 'tensorflow' has no attribute 'float32' #408

Open
Coloradohusky opened this issue Jun 10, 2023 · 1 comment
Open

Comments

@Coloradohusky
Copy link

System Information

  • Windows 10 Version 22H2 (OS Build 19045.2965
  • Python Version 3.7.9
  • TensorFlow-DirectML Version 1.15.8
  • AMD Radeon RX580 Version 22.10.20-220524a-380149C-AMD-Software-PRO-Edition
Host System
--------------------------------------------------------------------------------
Windows 10 Version  : Windows 10 Education 64-bit (10.0, Build 19045) (19041.vb_release.191206-1406)
Processor           : AMD Ryzen 5 2600 Six-Core Processor             (12 CPUs), ~3.4GHz
Memory              : 49152MB RAM
DirectX Version     : DirectX 12

Python Environment
--------------------------------------------------------------------------------
Python Version      : 3.7.9
TensorFlow-DirectML : 1.15.8

DirectX Device
--------------------------------------------------------------------------------
Description         : Radeon RX 580 Series
Manufacturer        : Advanced Micro Devices, Inc.
Chip Type           : AMD Radeon Graphics Processor (0x67DF)
Dedicated Memory    : 8170 MB
Driver Version      : 30.0.21020.2
Driver Model        : WDDM 2.7
Driver Date         : 5/23/2022 5:00:00 PM
Feature Levels      : 12_0,11_1,11_0,10_1,10_0,9_3,9_2,9_1

Repro Details

Describe the current behavior
Installing tensorflow-directml on a fresh venv, and running import tensorflow.compat.v1 as tf each time, as per https://learn.microsoft.com/en-us/windows/ai/directml/gpu-tensorflow-windows.
First issue was ModuleNotFoundError: No module named 'matplotlib', fixed with pip install matplotlib.
Second issue was ModuleNotFoundError: No module named 'IPython', fixed with pip install ipython.
Third issue was ModuleNotFoundError: No module named 'object_detection', so I ran pip install tensorflow-object-detection-api. This caused:

tensorflow-directml 1.15.8 requires numpy<1.19.0,>=1.16.0, but you have numpy 1.21.6 which is incompatible.
tensorflow-directml 1.15.8 requires tensorboard<1.16.0,>=1.15.0, but you have tensorboard 2.11.2 which is incompatible.
tensorflow-directml 1.15.8 requires tensorflow-estimator==1.15.1, but you have tensorflow-estimator 2.11.0 which is incompatible.

Final issue I ran into was AttributeError: module 'tensorflow' has no attribute 'float32', which I am unsure how to fix. I ran pip install tensorflow-directml again to see if that fixed anything, but it had no effect. I also tried pip install --upgrade --force-reinstall tensorflow and pip install --upgrade --force-reinstall tensorflow-directml, but they also had no effect.
Describe the expected behavior
import tensorflow.compat.v1 as tf runs with no errors.
Other info / logs

Python 3.7.9 (tags/v3.7.9:13c94747c7, Aug 17 2020, 18:58:18) [MSC v.1900 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow.compat.v1 as tf
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "C:\Users\Riley\Desktop\Misc stuff\tensorflow\tensorflow.py", line 15, in <module>
    from object_detection.utils import ops as utils_ops
  File "C:\Users\Riley\Desktop\Misc stuff\tensorflow\venv\lib\site-packages\object_detection\utils\ops.py", line 291, in <module>
    dtype=tf.float32):
AttributeError: module 'tensorflow' has no attribute 'float32'
@Petros626
Copy link

In my uninstalling tensorflow directml deactivating my venv, restart and then installing tensorflow directml in the venv, helped

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants