feedzai / feedzai-openml-java

Implementations for Feedzai's OpenML APIs to allow for usage of machine learning models in the Java programming language.
https://www.feedzai.com
Apache License 2.0
2 stars 11 forks source link

Add H2O GLM mandatory properties with default values for tests #48

Closed shengwangsw closed 4 years ago

shengwangsw commented 4 years ago

There are three mandatory parameters that H2O GLM 3.22.1.X doesn't use and H2O GLM 3.30.0.X does:

This commit adds these three parameters with its default values.

Added UT to ensure that all algorithms have the mandatory parameters

With the above UT, we discover that there are four mandatory parameters that H2O Isolation Forest 3.22.1.X already provided and we didn't add it.

This commit also adds these four parameters with its default values.

codecov[bot] commented 4 years ago

Codecov Report

Merging #48 into master will increase coverage by 0.37%. The diff coverage is n/a.

Impacted file tree graph

@@             Coverage Diff              @@
##             master      #48      +/-   ##
============================================
+ Coverage     77.38%   77.76%   +0.37%     
- Complexity      355      356       +1     
============================================
  Files            39       39              
  Lines          1318     1318              
  Branches        120      120              
============================================
+ Hits           1020     1025       +5     
+ Misses          230      224       -6     
- Partials         68       69       +1     
Impacted Files Coverage Δ Complexity Δ
...n/java/com/feedzai/openml/h2o/H2OModelCreator.java 73.41% <0.00%> (+6.32%) 16.00% <0.00%> (+1.00%)

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 ad56167...6dc19a1. Read the comment docs.

MiguelSCruz commented 4 years ago

I think your second link should point to http://docs.h2o.ai/h2o/latest-stable/h2o-docs/data-science/glm.html

shengwangsw commented 4 years ago

I think your second link should point to http://docs.h2o.ai/h2o/latest-stable/h2o-docs/data-science/glm.html

I don't know if you changed it, but it is pointing to the link you are mentioning

paulojrp commented 4 years ago

The new properties are explained in this link: https://github.com/h2oai/h2o-3/blob/jenkins-3.30.0.7/h2o-docs/src/product/data-science/glm.rst

MiguelSCruz commented 4 years ago

I think your second link should point to http://docs.h2o.ai/h2o/latest-stable/h2o-docs/data-science/glm.html

I don't know if you changed it, but it is pointing to the link you are mentioning

I didn't. Notice it's pointing to XGBoost and not GLM.

shengwangsw commented 4 years ago

I think your second link should point to http://docs.h2o.ai/h2o/latest-stable/h2o-docs/data-science/glm.html

I don't know if you changed it, but it is pointing to the link you are mentioning

I didn't. Notice it's pointing to XGBoost and not GLM.

Ah, you are right. I focused on the version but didn't confirm the algorithm. Thank you for noticing it