0

I am trying to use the Tensorflow object detection API to create my own model, so I created a virtudal enviorement,installed tensorflow 2.16.1, and I am using python 3.12.3. I installed Slim as well, and I want execute the proto files using protoc. When I run the command python model_builder_tf2_test.py to verify the setup and configuration of the TensorFlow Object Detection API with TensorFlow, it always generates an error.

File "C:\Users\user\tfodV\Lib\site-packages\object_detection\models\faster_rcnn_pnas_feature_extractor.py", line 24, in from nets.nasnet import nasnet_utils ModuleNotFoundError: No module named 'nets.nasnet'

When I verified the existence of the package, it does exist, I also added from the the object_detection folder the research folder, Slim folder and the nets to environment variable to maybe solve the issues yet the same problem arose.

Improve this question
2
  • This is an issue that you can find relevant. I am not familiar with tenserflow but it seems that nasnet is a submodule of nets. Sot this can help: github.com/tensorflow/models/issues/729 Commented Jun 17, 2024 at 14:42
  • Hi @wiem , The error you are facing is that the tf object Detection API cannot locate slim’s model components. Try to directly import path check in model builder_tf2_test.py.open that file and ensure nasnet_utils imports correctly import sys ,sys.path.append(‘slim_folder_path’).the re-run the test command. Commented Oct 31, 2024 at 11:29
Related questions
13112
What does the "yield" keyword do in Python?
8115
Does Python have a ternary conditional operator?
7519
What are metaclasses in Python?
Load 7 more related questions Show fewer related questions

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.