Training object detection model from scratch with tensorflow

Since I’ve just started working in this field, I’m getting lost among the resources. I encounter a lot of errors due to incompatibilities in the resources I follow. Is there a resource where I can access the steps in bulk?

It depends mainly on the sequence of steps you take to complete the task. Machine learning mastery is one of the good resource, especially for CNN and some basic deep learning stuff. I learned a lot from here and in my opinion if you done the basics correct then its all about coding it and tensorflow documentation will help you with that. There are lots of example colab notebooks as well which can headstart your understanding towards tensorflow. Hope this helps you. Happy learning

1 Like

Hi @abdullahsuheyl ,

Welcome to TF Forum!

Here are some resources that offer clear and comprehensive steps to guide you through the process:

The below one is great tutorial that can explain step by step in details.

  1. TensorFlow 2 Object Detection API tutorial — TensorFlow 2 Object Detection API tutorial documentation
  2. The TensorFlow Model Garden is a repository with a number of different implementations of state-of-the-art (SOTA) models and modeling solutions for TensorFlow users.
  3. Here couple of notebooks related to Image classification, Object Detection, Instance and semantic segmentation tutorials are available with beautiful explanation.
    4.If you face any issues while implementing or training , you can raise an issue in Model Garden Github repo where lot of folks can help you to resolve as well. For login new issue in github follow this template for easy access.
    5.If you wants to contribute any code to tensorflow model garden then you can follow this template.

Hope this helps!