Motivation

When creating a custom environment, deciding the correct TensorSpec (TS) to use is not straightforward. Some TS are never used, some look very similar to each other and sometimes docstrings are too concise or they miss some parameters.

Solution

After a brief talk with @vmoens and @fedebotu, we propose:

1. Improving the documentation: better detail docstrings and maybe add a tutorial for explaining what TS to use for different use cases.
2. Refactoring the tensor_specs module: review all the classes, their naming, their methods and check if there is a more intuitive way to achieve the same result.

Alternatives

The refactoring might happen as a v2 of the current tensor_specs module, to avoid breaking compatibility with previous versions of the code.

Additional context

This feature is proposed as a mean to improve user experience and as a possible follow up to #1795

Checklist

• I have checked that there is no similar issue in the repo (required)