Information Retrieval (3/5)

[lucadiliello]

Before submitting

• Was this discussed/approved via a Github issue? (no need for typos and docs improvements)
• Did you read the contributor guideline, Pull Request section?
• Did you make sure to update the docs?
• Did you write any new necessary tests?

What does this PR do?

Implements Precision (@k) for Information Retrieval. More info here.

PR review

Anyone in the community is free to review the PR once the tests have passed.
If we didn't discuss your PR in Github issues there's a high chance it will not be merged.

Did you have fun?

Make sure you had fun coding 🙃

[codecov]

Codecov Report

Merging #139 (fc75878) into master (165cff0) will increase coverage by 0.07%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #139      +/-   ##
==========================================
+ Coverage   96.68%   96.76%   +0.07%     
==========================================
  Files         152      156       +4     
  Lines        4762     4816      +54     
==========================================
+ Hits         4604     4660      +56     
+ Misses        158      156       -2     

Flag	Coverage Δ
Linux	79.27% <87.50%> (+0.06%)	⬆️
Windows	79.27% <87.50%> (+0.06%)	⬆️
cpu	96.76% <100.00%> (+0.07%)	⬆️
gpu	96.76% <ø> (+0.07%)	⬆️
macOS	96.76% <100.00%> (+0.07%)	⬆️
pytest	96.76% <100.00%> (+0.07%)	⬆️
python3.6	96.75% <100.00%> (+0.07%)	⬆️
python3.8	96.76% <100.00%> (+0.07%)	⬆️
python3.9	96.76% <100.00%> (+0.07%)	⬆️
torch1.3.1	95.54% <100.00%> (+0.09%)	⬆️
torch1.4.0	95.68% <100.00%> (+0.09%)	⬆️
torch1.8.1	96.76% <100.00%> (+0.07%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
...hmetrics/functional/retrieval/average_precision.py	100.00% <ø> (ø)
...rchmetrics/functional/retrieval/reciprocal_rank.py	100.00% <ø> (ø)
torchmetrics/__init__.py	100.00% <100.00%> (ø)
torchmetrics/functional/__init__.py	100.00% <100.00%> (ø)
torchmetrics/functional/retrieval/__init__.py	100.00% <100.00%> (ø)
torchmetrics/functional/retrieval/precision.py	100.00% <100.00%> (ø)
torchmetrics/retrieval/__init__.py	100.00% <100.00%> (ø)
torchmetrics/retrieval/mean_average_precision.py	100.00% <100.00%> (ø)
torchmetrics/retrieval/mean_reciprocal_rank.py	100.00% <100.00%> (ø)
torchmetrics/retrieval/retrieval_precision.py	100.00% <100.00%> (ø)
... and 11 more

---

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 165cff0...fc75878. Read the comment docs.

[lucadiliello]

Changed also the test of MRR to compare against the sklearn implementation instead of mine.
For Precision, I didn't find a reliable implementation :(

[SkafteNicki]

smaller comments else LGTM

[SkafteNicki]

@lucadiliello tests seems to be failing (missing import I think). Could you look into it? :]

[Borda]

Suggested change

	<https://en.wikipedia.org/wiki/Mean_reciprocal_rank>`__.
	<https://en.wikipedia.org/wiki/Mean_reciprocal_rank>`__.

is this the correct link formatting?

[lucadiliello]

I receive an error when doing make docs if I use a single underscore. This is probably caused by having the same link to wikipedia 2 times: 1 link in the Metric class and one in the functional version.
The error is:

/Users/lucadiliello/development/metrics/torchmetrics/functional/retrieval/reciprocal_rank.py:docstring of torchmetrics.functional.retrieval.reciprocal_rank.retrieval_reciprocal_rank:22: WARNING: Duplicate explicit target name: "here".

So I found that by using 2 underscores it does work.