[WIP]add examples using Nano HPO to tune the hyperparameters in tensorflow training

[Chaselzxy]

Description

Add 5 examples using Nano HPO to tune the hyperparameters in tensorflow training.

1. Why the change?

Divide the tutorial of how to use Nano HPO to tune the hyperparameters in tensorflow training into several small examples.

2. User API changes

3. Summary of the change

model arch search - using sequential API model arch search - using functional API model arch search - by subclassing keras.Model learning rate search batch size search

4. How to test?

test in tutorial

5. New dependencies

[shane-huang]

Each example should be a smallest runnable program which demonstrate just a single functionality. For example, for batch size search, you should only specify batch size for search (now you still have model arch search, i.e. search spaces in model build).

[shane-huang]

Replace spaces with “_” in example file names.

[shane-huang]

change these ipynb examples into simple .py examples. Remeber to remove the pip install parts.