add sign, sgn

[mtar]

Description

This PR adds the sign & sgn functions to heat. The former behaves like numpy on complex values, the latter like pytorch.

Note
Requires PyTorch 1.9 Uses a indexing mask on older versions now.

Issue/s resolved: #780

Changes proposed:

• add sign
• add sgn

Type of change

• New feature

Due Diligence

• All split configurations tested
• Multiple dtypes tested in relevant functions
• Documentation updated (if needed)
• Updated changelog.md under the title "Pending Additions"

Does this change modify the behaviour of other functions? If so, which?

no

[codecov]

Codecov Report

Merging #827 (5fba209) into master (d00d61e) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master     #827   +/-   ##
=======================================
  Coverage   95.47%   95.48%           
=======================================
  Files          64       64           
  Lines        9436     9454   +18     
=======================================
+ Hits         9009     9027   +18     
  Misses        427      427           

Flag	Coverage Δ
gpu	94.61% <100.00%> (+0.01%)	⬆️
unit	91.00% <94.73%> (+<0.01%)	⬆️

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

Impacted Files	Coverage Δ
heat/core/rounding.py	96.96% <100.00%> (+0.67%)	⬆️

---

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 d00d61e...5fba209. Read the comment docs.

[mtar]

test fails on pytorch 1.8.1: nonzero can't be used with complex dtypes

[coquelin77]

rerun tests

[coquelin77]

it looks like the tests are failing here because of device mismatches

[coquelin77]

test fails on pytorch 1.8.1: nonzero can't be used with complex dtypes

this seems to be the case for torch 1.7 as well

[mtar]

Tests updated for GPU.
Two missing lines in coverage because of code for pytorch 1.9 only.

[ClaudiaComito]

👏🏼 @mtar thanks a lot as usual. Just a few comments.