use pytest.approx insetad of np.array_allclose for a single float

[massich]

Reference Issue

#323

What does this implement/fix? Explain your changes.

This is just an Idea of how the test would be. I would also use fixture to create a matrix of parameters and expected values. (instead of using this for loop)

Any other comments?

[codecov]

Codecov Report

Merging #380 into master will increase coverage by 0.01%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master     #380      +/-   ##
==========================================
+ Coverage   98.71%   98.72%   +0.01%     
==========================================
  Files          68       67       -1     
  Lines        3967     3929      -38     
==========================================
- Hits         3916     3879      -37     
+ Misses         51       50       -1

Impacted Files	Coverage Δ
...er_sampling/prototype_generation/tests/__init__.py
imblearn/under_sampling/base.py	100% <0%> (+2.43%)	⬆️

---

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 14bda7c...946a2f5. Read the comment docs.

[glemaitre]

Missing:

• datasets/tests/test_zenodo.py
• metrics/tests/test_classification.py

[glemaitre]

This is just an Idea of how the test would be. I would also use fixture to create a matrix of parameters and expected values. (instead of using this for loop)

Yep but we need things to move quick