Example: Imagenet (Baselines)

[penguinmenac3]

Is your feature request related to a problem? Please describe.
Write a tutorial on using inception_v3 or vgg to classify imagenet images. This could help starters with image classificaton.

Describe the solution you'd like
An example like mnist that shows how to train vgg16 or inception_v3 on image net or a similar but smaller task. Ideally this would include transfer learning. Starting with a pretrained vgg or inception network and learning classification on a smaller dataset. For loading a small dataset that is stored in a folder per class the opendatalake.classification.named_folders package is ideal.

Describe alternatives you've considered
None

Additional context
The example should have a similar structure to the mnist example.
This is an ideal issue for someone who wants to start contributing.

An example contains:

• hyper_params.json
• loss.py
• prepare_training.py
• train.py