Maksym Sladkov

This *ngxtop* is cool. I obviously never heard of it. But one of the options would be to run it in a container itself [docker image](https://hub.docker.com/r/mishunika/ngxtop/) and then mount a...

A. I'm might be biased by knowing that WDF subclassing DF, but IO methods of DF is one of the cornerstones of this library. The legend says that it all...

A. What I wanted to say is if we keep the DF nature intact, then we should not shy away from using IO methods (https://pandas.pydata.org/pandas-docs/stable/io.html). If we treat WDF as...

You are right, it is indeed a factory method... If that is how `pandas` does it, then I absolutely don't want to invent anything new. 1. I don't have a...

1. Ok, because my alternative proposal would be to allow importing WDF and WDF returning `read_strava` method directly from the `sweat.io` namespace (basically what `pandas` does with their IO methods)...

I think it is ok, the `sweat.io.strava.api` is meant for brave developers and they should be able to discover it. btw, is it a normal practice to define functions in...

In general looks good. Splitting metrics into modules according to the observed variable poses a danger of duplicating similar calculations twice. One of the examples - `algorithms.streams.compute_zone()` can be applied...

I don't keep streams - they are a legacy of `vmpy`. The content of both `streams.py` and `metrics.py` will move into whatever we decide is good for a project structure....

Great you brought it up - I thought about exactly the same ;)

I also like Sphynx and Read the Docs. I'm currently using NumPy style docstring, but Sphynx support few more styles if we decide that NumPy is too long. The same...