1

I installed plaidML to utilize intel UHD graphics card for tensorflow. I ran the code below to check if the plaidML deep learning framework works well as a backend of Keras.

import plaidml.keras

plaidml.keras.install_backend()

Then the following error occurred: As a result of what I found on stack overflow, all of the error codes below were solved by replacing 'keras' with 'tensorflow.keras'. But in my case it didn't work. My version of tensorflow and keras are 2.5.0 and 2.2.4 respectively.

    2022-01-29 21:22:29.043422: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'cudart64_110.dll'; dlerror: cudart64_110.dll not found
    2022-01-29 21:22:29.043619: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
    Traceback (most recent call last):
      File "G:/shared drive/Algorithm/Reinf trading/test_plaidML.py", line 3, in <module>
        plaidml.keras.install_backend()
      File "C:\Users\VsL\.conda\envs\tf25_py38\lib\site-packages\plaidml\keras\__init__.py", line 68, in install_backend
        from keras.utils import conv_utils
      File "C:\Users\VsL\.conda\envs\tf25_py38\lib\site-packages\keras\__init__.py", line 3, in <module>
        from . import utils
      File "C:\Users\VsL\.conda\envs\tf25_py38\lib\site-packages\keras\utils\__init__.py", line 27, in <module>
        from .multi_gpu_utils import multi_gpu_model
      File "C:\Users\VsL\.conda\envs\tf25_py38\lib\site-packages\keras\utils\multi_gpu_utils.py", line 7, in <module>
        from ..layers.merge import concatenate
      File "C:\Users\VsL\.conda\envs\tf25_py38\lib\site-packages\keras\layers\__init__.py", line 4, in <module>
        from ..engine.base_layer import Layer
      File "C:\Users\VsL\.conda\envs\tf25_py38\lib\site-packages\keras\engine\__init__.py", line 3, in <module>
        from .input_layer import Input
      File "C:\Users\VsL\.conda\envs\tf25_py38\lib\site-packages\keras\engine\input_layer.py", line 7, in <module>
        from .base_layer import Layer
      File "C:\Users\VsL\.conda\envs\tf25_py38\lib\site-packages\keras\engine\base_layer.py", line 12, in <module>
        from .. import initializers
      File "C:\Users\VsL\.conda\envs\tf25_py38\lib\site-packages\keras\initializers\__init__.py", line 124, in <module>
        populate_deserializable_objects()
      File "C:\Users\VsL\.conda\envs\tf25_py38\lib\site-packages\keras\initializers\__init__.py", line 82, in populate_deserializable_objects
        generic_utils.populate_dict_with_module_objects(
    AttributeError: module 'keras.utils.generic_utils' has no attribute 'populate_dict_with_module_objects'
    
    Process finished with exit code 1

Thank you in advance

VirusVersa
  • 11
  • 2
  • This is most common issue with `Keras=2.2.4`. Use Keras>=2.3 or replace keras Api's with tf.keras. `from tensorflow.keras.utils import_____`. Thanks! –  Feb 01 '22 at 09:37

0 Answers0