Regression experiment documentation

[amallia]

Added regression experiment for Robust04

[elshize]

A few comments but here's an idea that we've kind of touched on with @JMMackenzie before. How about create a docker image that executes the entire script? You'd have to mount the input directory with the collection, but then all else would just work automatically. This way, we can very quickly run these tests ourselves.

[amallia]

A few comments but here's an idea that we've kind of touched on with @JMMackenzie before. How about create a docker image that executes the entire script? You'd have to mount the input directory with the collection, but then all else would just work automatically. This way, we can very quickly run these tests ourselves.

I am a bit against Docker for this. It is good if we want to automatically keep updated, for example, effectiveness results on several corpuses. I think having a recipe-stile guide is still good in order to document how to use PISA.

Also, docker is not always great if you want to use it in interactive mode.

[elshize]

I am not against the guide, it's good to have it, for sure. But I think docker image also would be nice.

[JMMackenzie]

Looks good to me. I am agnostic on the Docker issue. Could always come back to it if we like?

[elshize]

Yeah, we can do (or not) docker later, just saying...

[amallia]

@elshize @JMMackenzie I have updated the documentation according to your comments.

[codecov]

Codecov Report

Merging #354 into master will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #354   +/-   ##
=======================================
  Coverage   92.41%   92.41%           
=======================================
  Files          90       90           
  Lines        4915     4915           
=======================================
  Hits         4542     4542           
  Misses        373      373           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 15ed0d6...b397caf. Read the comment docs.

[amallia]

@elshize addressed comments

[elshize]

@amallia I added more formatting fixes, take a look if okay.