Image Classification - Rooms


I’m “forced” to create a model for image classification with Keras which should predict three specific rooms in my organisation.
Sadly I didn’t learn/work with Python/AI and Keras before and am a bit overwhelmed (I mostly work with C++).
I already set up a jupyer training environment on my server and created a test model with the CatsVSDogs, which worked without problems.
Now to predict three rooms, I know I have to define labels for the three rooms.
I looked into the code examples: Code examples and there are many different types of image classifications with different methods.
Can someone recommend me a tutorial or code example which is the most applicable for my case?
And is there a way to know how many pictures per rooms are recommended?


Please follow this tutorial for multi class classification use cases using Tensorflow. For image classification using deep learning, a rule of thumb is 1,000 images per class where this number can go down significantly if one uses pre-trained models. Thank you.