Accuracy average

[arv-77]

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?

Fixes #99
Closes #155
I have added the average argument with all other required arguments. For subset_accuracy I have retained some sections of the previous code.

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 🙃

[pep8speaks]

Hello @arvindmuralie77! Thanks for updating this PR.

There are currently no PEP 8 issues detected in this Pull Request. Cheers! 🍻

Comment last updated at 2021-04-16 13:28:34 UTC

[Borda]

@arvindmuralie77 how is it going here, do you think we can squeeze it to 0.3.0?

[arv-77]

do you think we can squeeze it to 0.3.0?

Dear @Borda , what exactly mean by this?

[Borda]

@arvindmuralie77 I was just asking what is you estimate to finish this PR :]

[codecov]

Codecov Report

Merging #166 (7457f54) into master (953b621) will decrease coverage by 0.17%.
The diff coverage is 97.72%.

@@            Coverage Diff             @@
##           master     #166      +/-   ##
==========================================
- Coverage   96.18%   96.01%   -0.18%     
==========================================
  Files         180       90      -90     
  Lines        5582     2859    -2723     
==========================================
- Hits         5369     2745    -2624     
+ Misses        213      114      -99     

Flag	Coverage Δ
Linux	79.18% <68.18%> (-0.61%)	⬇️
Windows	79.18% <68.18%> (-0.61%)	⬇️
cpu	96.01% <97.72%> (-0.16%)	⬇️
gpu	?
macOS	96.01% <97.72%> (-0.16%)	⬇️
pytest	96.01% <97.72%> (-0.18%)	⬇️
python3.6	95.99% <97.72%> (-0.16%)	⬇️
python3.8	96.01% <97.72%> (-0.16%)	⬇️
python3.9	95.90% <97.72%> (-0.16%)	⬇️
torch1.3.1	94.94% <97.72%> (-0.13%)	⬇️
torch1.4.0	95.06% <97.72%> (-0.14%)	⬇️
torch1.8.1	95.90% <97.72%> (-0.16%)	⬇️

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

Impacted Files	Coverage Δ
torchmetrics/utilities/checks.py	92.43% <80.00%> (-0.35%)	⬇️
torchmetrics/functional/classification/accuracy.py	93.93% <97.95%> (-6.07%)	⬇️
torchmetrics/classification/accuracy.py	96.15% <100.00%> (-3.85%)	⬇️
...torchmetrics/functional/classification/accuracy.py
...etrics/functional/regression/mean_squared_error.py
.../1/s/torchmetrics/retrieval/retrieval_precision.py
...s/torchmetrics/classification/average_precision.py
...trics/functional/regression/mean_absolute_error.py
...rics/functional/classification/confusion_matrix.py
... and 84 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 953b621...7457f54. Read the comment docs.

[arv-77]

Dear @Borda , I didn't know there was a deadline for 0.3.0. Everything is complete, if default value of mdmc_average is set as global, then the code is passing on all the previously existing test cases. Now only new test cases need to be created since new parameters have been added to accuracy.

[Borda]

I didn't know there was a deadline for 0.3.0

there is no deadline, I just thought that it would be nice to have it there, so just asking how do you feel about it :]

[arv-77]

Hey @Borda , I have added some new test cases for the updated class. I was having some issues understanding the multi-dimension-multilabel class. Apart from that, I think I have covered all other cases. Should I convert this draft to a PR so that you can review it.

[SkafteNicki]

smaller comments

[arv-77]

@SkafteNicki , I have corrected the mentioned issues.

[arv-77]

Done @Borda :]

[Borda]

a few more tests please and we are ready to land :]

[arv-77]

I have added more test cases as mentioned