Object Detection with Tensorflow 2.17

Hello!

I am new to tensorflow and I am working on creating an object detection model. I am following along with a video found on youtube linked here:

[https://www.youtube.com/watch?v=yqkISICHH-U]

A walkthrough tutorial along with the code I have been using can be found here:

[GitHub - nicknochnack/TFODCourse]

I am finding that as I am going through the tutorial I am coming across a lot of issues in regards to the migration to tensorflow 2. Many of them related to tf.contrib which I have manually fixed on my own. However, there are some that are just extremely tedious and I do not know how to fix them. For example, I am now currently running into the the following error:

ModuleNotFoundError: No module named 'tensorflow_io'

When I try to install tensorflow_io from the command prompt in my virtual environment, I get the following error message.

ERROR: Could not find a version that satisfies the requirement tensorflow_io (from versions: none)
ERROR: No matching distribution found for tensorflow_io

I tried reverting back to tensor flow version 2.16 but this still did not resolve the issue.

So all in all, I have the following questions.

  1. Does anyone know how to solve the tensorflow_io installation issue above?
  2. Is the tutorial that I am following outdated?
  3. If so, can anyone point me to another resource I could use to help train an object detection model?

Thank you in advance for any advice that you can spare!

Hi @Johnny_Donza, Welcome to tensorflow community! We’re glad to have you here.

We are able to install this tensorflow_io module with tensorflow version 2.17 in colab. Could you please let us know which OS you have installed, the environment you are trying to execute your code?

The model ssd_mobilenet_v2_320x320_coco17_tpu-8.config is research model and it might have compatibility issue with latest tensorflow versions. I recommend you to follow this latest tutorial .

Thank you! We are happy to help you.