-
Notifications
You must be signed in to change notification settings - Fork 2k
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
add R unit test to check AIC, loglikelihood comparison to R.
- Loading branch information
1 parent
797b136
commit d4708df
Showing
1 changed file
with
75 additions
and
0 deletions.
There are no files selected for viewing
75 changes: 75 additions & 0 deletions
75
h2o-r/tests/testdir_algos/glm/runit_PUBDEV_8947_GLM_likelihoods_aic.R
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,75 @@ | ||
| setwd(normalizePath(dirname(R.utils::commandArgs(asValues=TRUE)$"f"))) | ||
| source("../../../scripts/h2o-r-test-setup.R") | ||
|
|
||
| test.glm.aic.likelihood <- function() { | ||
| # checking for Gaussian | ||
| h2o.data = h2o.uploadFile(locate("smalldata/prostate/prostate_complete.csv.zip"), destination_frame="h2o.data") | ||
| R.data = as.data.frame(as.matrix(h2o.data)) | ||
| myY = "GLEASON" | ||
| myX = c("ID","AGE","RACE","CAPSULE","DCAPS","PSA","VOL","DPROS") | ||
| R.formula = (R.data[,"GLEASON"]~.) | ||
| model.h2o.gaussian.identity <- h2o.glm(x=myX, y=myY, training_frame=h2o.data, family="gaussian", link="identity",alpha=0.5, lambda=0, nfolds=0) | ||
| model.R.gaussian.identity <- glm(formula=R.formula, data=R.data[,2:9], family=gaussian(link=identity), na.action=na.omit) | ||
| perf <- h2o.performance(model.h2o.gaussian.identity, h2o.data) | ||
| print("GLM Gaussian") | ||
| print("H2O AIC") | ||
| print(h2o.aic(perf)) | ||
| print("H2O log likelihood") | ||
| # print(h2o.loglikelihood(perf)) # Yuliia: please implement this | ||
| print("R AIC") | ||
| print(AIC(model.R.gaussian.identity)) | ||
| print("R loglikelihood") | ||
| print(logLik(model.R.gaussian.identity)) | ||
| # browser() | ||
|
|
||
| # check for Gamma | ||
| myY = "DPROS" | ||
| myX = c("ID","AGE","RACE","CAPSULE","DCAPS","PSA","VOL","GLEASON") | ||
| R.formula = (R.data[,"DPROS"]~.) | ||
| model.h2o.gamma.inverse <- h2o.glm(x=myX, y=myY, training_frame=h2o.data, family="gamma", link="inverse",alpha=0.5, lambda=0, nfolds=0) | ||
| model.R.gamma.inverse <- glm(formula=R.formula, data=R.data[,c(1:5,7:9)], family=Gamma(link=inverse), na.action=na.omit) | ||
| perf <- h2o.performance(model.h2o.gamma.inverse, h2o.data) | ||
| print("GLM Gamma") | ||
| print("H2O AIC") | ||
| print(h2o.aic(perf)) | ||
| print("H2O log likelihood") | ||
| # print(h2o.loglikelihood(perf)) # Yuliia: please implement this | ||
| print("R AIC") | ||
| print(AIC(model.R.gamma.inverse)) | ||
|
|
||
| # checking for binomial | ||
| myY <- "CAPSULE" | ||
| myX <- c("AGE","RACE","DCAPS","PSA","VOL","DPROS","GLEASON") | ||
| R.formula <- (R.data[,"CAPSULE"]~.) | ||
| model.h2o.binomial.logit <- h2o.glm(x=myX, y=myY, training_frame=h2o.data, family="binomial", link="logit",alpha=0.5, lambda=0, nfolds=0) | ||
| model.R.binomial.logit <- glm(formula=R.formula, data=R.data[,4:10], family=binomial(link=logit), na.action=na.omit) | ||
| perf <- h2o.performance(model.h2o.binomial.logit, h2o.data) | ||
| print("GLM binomial") | ||
| print("H2O AIC") | ||
| print(h2o.aic(perf)) | ||
| print("H2O log likelihood") | ||
| # print(h2o.loglikelihood(perf)) # Yuliia: please implement this | ||
| print("R AIC") | ||
| print(AIC(model.R.binomial.logit)) | ||
| print("R loglikelihood") | ||
| print(logLik(model.R.binomial.logit)) | ||
|
|
||
| # checking for Poisson | ||
| myY = "GLEASON" | ||
| myX = c("ID","AGE","RACE","CAPSULE","DCAPS","PSA","VOL","DPROS") | ||
| R.formula = (R.data[,"GLEASON"]~.) | ||
| model.h2o.poisson.log <- h2o.glm(x=myX, y=myY, training_frame=h2o.data, family="poisson", link="log",alpha=0.5, lambda=0, nfolds=0) | ||
| model.R.poisson.log <- glm(formula=R.formula, data=R.data[,2:9], family=poisson(link=log), na.action=na.omit) | ||
| perf <- h2o.performance(model.h2o.poisson.log, h2o.data) | ||
| print("GLM Poisson") | ||
| print("H2O AIC") | ||
| print(h2o.aic(perf)) | ||
| print("H2O log likelihood") | ||
| # print(h2o.loglikelihood(perf)) # Yuliia: please implement this | ||
| print("R AIC") | ||
| print(AIC(model.R.poisson.log)) | ||
| print("R loglikelihood") | ||
| print(logLik(model.R.poisson.log)) | ||
| } | ||
|
|
||
| doTest("Testing AIC/likelihood for GLM", test.glm.aic.likelihood) |