[#43] probabilities

[Samuel-Campbell]

1) Added probabilities to predictions 2) Model now trains using probabilities = True --> increases training time 3) ReadMe updated with change in end point 4) Probabilities only on Classifiers obviously.

Only once approved

• [ ] upload new models on cjl

[codecov[bot]]

Codecov Report

Merging #390 into master will increase coverage by 0.13%. The diff coverage is 83.33%.

@@            Coverage Diff             @@
##           master     #390      +/-   ##
==========================================
+ Coverage   68.08%   68.21%   +0.13%     
==========================================
  Files          51       51              
  Lines        2074     2089      +15     
  Branches       17       17              
==========================================
+ Hits         1412     1425      +13     
- Misses        657      659       +2     
  Partials        5        5

Impacted Files	Coverage Δ
...raining/classifier/multi_output/multi_class_svm.py	87.76% <100%> (+0.64%)	:arrow_up:
src/ml_service/web/ml_controller.py	63.23% <66.66%> (+1.56%)	:arrow_up:

---

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 6c6a1a5...8c57034. Read the comment docs.

[Samuel-Campbell]

Everyone telling me mixed stuff but friday we discussed this. When we do a prediction we wanted to give the user it's probability. Answering more questions will increase probability or change the outcome

[Samuel-Campbell]

Also what was said: "just give us back as many numbers as you can"