h2o-3
h2o-3 copied to clipboard
h2oai
H2O is an Open Source, Distributed, Fast & Scalable Machine Learning Platform: Deep Learning, Gradient Boosting (GBM) & XGBoost, Random Forest, Generalized Linear Modeling (GLM with Elastic Net), K-Me...
This PR completes the task in JIRA: https://h2oai.atlassian.net/browse/PUBDEV-8685 . I have completed the implementation in Java. I need to generate tweedie dataset so that I can generate comprehensive tests to...
This PR completes the task in JIRA: https://h2oai.atlassian.net/browse/PUBDEV-8574 For dispersion parameter estimation, apart from ml and pearson mode, I added the deviance mode. In addition, I added tests in Java,...
This PR completes the task in JIRA: https://h2oai.atlassian.net/browse/PUBDEV-8786 Added predictors added/removed per step to modelselection toolbox in the result frame and the model summary. Added client functions to access the...
Set order method to radix as the default in 3.3 does not support multiple decreasing criteria. **TODO**: * remove R3.3 from PR stage
To review, compare against the [Jetty implementation](https://raw.githubusercontent.com/eclipse/jetty.project/jetty-9.4.40.v20210413/jetty-server/src/main/java/org/eclipse/jetty/server/Response.java). This brings things current to the version of Jetty being imported. https://h2oai.atlassian.net/browse/PUBDEV-8802
https://h2oai.atlassian.net/browse/PUBDEV-8315 Restoring the full model rendering for default output or `print(model)` in iPython/Jupyter or Py REPL
Move it to Model, rename, and use it also in XGboost for scoreContribution logic..
This PR completes the task in JIRA: https://h2oai.atlassian.net/browse/PUBDEV-8785 Instead of trying to optimize memory use, I use the full predictor size here. It seems that for small number of predictors,...