def testGLMParameters(prostateDataset):
features = ['AGE', 'RACE', 'DPROS', 'DCAPS', 'PSA']
algorithm = H2OGLM(seed=1, labelCol="CAPSULE", alphaValue=[0.5], lambdaValue=[0.5], maxIterations=30,
objectiveEpsilon=0.001, gradientEpsilon=0.001, objReg=0.001, maxActivePredictors=3000,
lambdaMinRatio=0.001, featuresCols=features)
model = algorithm.fit(prostateDataset)
compareParameterValues(algorithm, model)
def createInitialGlmDefinitionForRandomCols():
return H2OGLM(featuresCols=["x1", "x3", "x5", "x6"],
labelCol="y",
family="gaussian",
randomFamily=["gaussian"],
randomLink=["identity"],
HGLM=True,
calcLike=True)
def createInitialGlmDefinition():
featuresCols = [
"economy", "displacement", "power", "weight", "acceleration",
"year", "economy_20mpg"
]
return H2OGLM(featuresCols=featuresCols,
labelCol="cylinders",
seed=1,
splitRatio=0.8)
def testPropagationOfPredictionCol(prostateDataset):
predictionCol = "my_prediction_col_name"
algo = H2OGLM(featuresCols=["CAPSULE", "RACE", "DPROS", "DCAPS", "PSA", "VOL", "GLEASON"],
labelCol="AGE",
seed=1,
splitRatio=0.8,
predictionCol=predictionCol)
model = algo.fit(prostateDataset)
columns = model.transform(prostateDataset).columns
assert True == (predictionCol in columns)
def testPipelineSerialization(prostateDataset):
algo = H2OGLM(featuresCols=["CAPSULE", "RACE", "DPROS", "DCAPS", "PSA", "VOL", "GLEASON"],
labelCol="AGE",
seed=1,
splitRatio=0.8)
pipeline = Pipeline(stages=[algo])
pipeline.write().overwrite().save("file://" + os.path.abspath("build/glm_pipeline"))
loadedPipeline = Pipeline.load("file://" + os.path.abspath("build/glm_pipeline"))
model = loadedPipeline.fit(prostateDataset)
model.write().overwrite().save("file://" + os.path.abspath("build/glm_pipeline_model"))
loadedModel = PipelineModel.load("file://" + os.path.abspath("build/glm_pipeline_model"))
loadedModel.transform(prostateDataset).count()
def testInteractionColumnNamesArePassedWithoutException(spark):
data = [(0.0, "a", 2.0), (float("nan"), "b", 8.0), (0.0, "a", 4.0),
(1.0, "b", 1.0)]
df = spark.createDataFrame(data, ["x", "y", "z"])
plugValues = {"x": 0, "x_y.a": 1, "x_y.b": 2, "y": "b"}
glm = H2OGLM(labelCol="z",
seed=42,
ignoreConstCols=False,
standardize=False,
family="gaussian",
missingValuesHandling="PlugValues",
plugValues=plugValues)
glm.fit(df)
def testPipelineSerializationGLM(prostateDataset):
gridSearchTester(H2OGLM().setLabelCol("AGE"), prostateDataset)
def testPipelineSerializationGLM(prostateDataset):
gridSearchTester(H2OGLM(), prostateDataset)
def testParams():
glm = H2OGLM(modelId=None,
splitRatio=1.0,
labelCol="label",
weightCol=None,
featuresCols=[],
allStringColumnsToCategorical=True,
columnsToCategorical=[],
nfolds=0,
keepCrossValidationPredictions=False,
keepCrossValidationFoldAssignment=False,
parallelizeCrossValidation=True,
seed=-1,
distribution="AUTO",
convertUnknownCategoricalLevelsToNa=False,
standardize=True,
family="gaussian",
link="family_default",
solver="AUTO",
tweedieVariancePower=0.0,
tweedieLinkPower=0.0,
alpha=[1],
lambda_=None,
missingValuesHandling="MeanImputation",
prior=-1.0,
lambdaSearch=False,
nlambdas=-1,
nonNegative=False,
exactLambdas=False,
lambdaMinRatio=-1.0,
maxIterations=-1,
intercept=True,
betaEpsilon=1e-4,
objectiveEpsilon=-1.0,
gradientEpsilon=-1.0,
objReg=-1.0,
computePValues=False,
removeCollinearCols=False,
interactions=None,
interactionPairs=None,
earlyStopping=True,
foldCol=None,
predictionCol="prediction",
detailedPredictionCol="detailed_prediction",
withDetailedPredictionCol=False,
convertInvalidNumbersToNa=False)
assert glm.getModelId() == None
assert glm.getSplitRatio() == 1.0
assert glm.getLabelCol() == "label"
assert glm.getWeightCol() == None
assert glm.getFeaturesCols() == []
assert glm.getAllStringColumnsToCategorical() == True
assert glm.getColumnsToCategorical() == []
assert glm.getNfolds() == 0
assert glm.getKeepCrossValidationPredictions() == False
assert glm.getKeepCrossValidationFoldAssignment() == False
assert glm.getParallelizeCrossValidation() == True
assert glm.getSeed() == -1
assert glm.getDistribution() == "AUTO"
assert glm.getConvertUnknownCategoricalLevelsToNa() == False
assert glm.getStandardize() == True
assert glm.getFamily() == "gaussian"
assert glm.getLink() == "family_default"
assert glm.getSolver() == "AUTO"
assert glm.getTweedieVariancePower() == 0.0
assert glm.getTweedieLinkPower() == 0.0
assert glm.getAlpha() == [1.0]
assert glm.getLambda() == None
assert glm.getMissingValuesHandling() == "MeanImputation"
assert glm.getPrior() == -1.0
assert glm.getLambdaSearch() == False
assert glm.getNlambdas() == -1
assert glm.getNonNegative() == False
assert glm.getExactLambdas() == False
assert glm.getLambdaMinRatio() == -1.0
assert glm.getMaxIterations() == -1
assert glm.getIntercept() == True
assert glm.getBetaEpsilon() == 1e-4
assert glm.getObjectiveEpsilon() == -1.0
assert glm.getGradientEpsilon() == -1.0
assert glm.getObjReg() == -1.0
assert glm.getComputePValues() == False
assert glm.getRemoveCollinearCols() == False
assert glm.getInteractions() == None
assert glm.getInteractionPairs() == None
assert glm.getEarlyStopping() == True
assert glm.getFoldCol() == None
assert glm.getPredictionCol() == "prediction"
assert glm.getDetailedPredictionCol() == "detailed_prediction"
assert glm.getWithDetailedPredictionCol() == False
assert glm.getConvertInvalidNumbersToNa() == False
def createInitialGlmDefinition():
return H2OGLM(featuresCols=featuresCols, labelCol="CAPSULE", seed=1, splitRatio=0.8)