DeepMIMO Forum

Clear all

[Solved] Error in implementation of "Deep Learning Coordinated Beamforming for Highly-Mobile Millimeter Wave Systems"  


New Member
Joined: 12 months ago
Posts: 1
26/12/2022 11:29 am  


I am trying to recreate the figures of  "Deep Learning Coordinated Beamforming for Highly-Mobile Millimeter Wave Systems" paper. 

I am running the code in Spyder in Anaconda from a Windows 10 PC.
I am facing some errors with the python code like:

ValueError: Dimensions must be equal, but are 512 and 0 for '{{node mean_squared_error/SquaredDifference}} = SquaredDifference[T=DT_FLOAT](sequential_7/dense_47/Relu, IteratorGetNext:1)' with input shapes: [?,512], [?,0].


2022-12-21 12:35:51.507369: W tensorflow/core/util/] Could not open DLCB_code_output\Results_mmWave_ML_BSBS0: UNKNOWN: NewRandomAccessFile failed to Create/Open: DLCB_code_output\Results_mmWave_ML_BSBS0 : Access is denied.
; Input/output error

I would like to ask you if you are aware of the errors, and if so, what is the solution to the problem.
Is there a new version of the python script that works properly ?


Thank you very much.

Member Admin
Joined: 2 years ago
Posts: 39
28/12/2022 12:09 am  

Hi @gkonstan,


The codes in the Github repositories are up-to-date and work as is. We are not aware of any error so your problem may be due to the Python/TensorFlow versions you have (works with Python 3.7 and corresponding TensorFlow).

As a workaround, you may try the full python implementation, which does not include the operation you received the error (saving/loading the models).


Best regards,


This post was modified 11 months ago by udemirhan


Leave a reply

Author Name

Author Email

Title *

Maximum allowed file size is 10MB

Preview 0 Revisions Saved