def testPropagationOfPredictionCol(prostateDataset):
predictionCol = "my_prediction_col_name"
algo = H2OGAM(featuresCols=["DPROS", "DCAPS", "RACE", "GLEASON"],
gamCols=["PSA", "AGE"],
labelCol="CAPSULE",
seed=1,
splitRatio=0.8,
predictionCol=predictionCol)
model = algo.fit(prostateDataset)
columns = model.transform(prostateDataset).columns
assert True == (predictionCol in columns)
def testH2OGAMRegressorBehavesTheSameAsGenericH2OGAMOnNumericLabelColumn(
prostateDataset):
[trainingDateset, testingDataset] = prostateDataset.randomSplit([0.9, 0.1],
42)
automl = setParamtersForProblemSpecificTests(H2OGAM())
referenceModel = automl.fit(trainingDateset)
referenceDataset = referenceModel.transform(testingDataset)
classifier = setParamtersForProblemSpecificTests(H2OGAMRegressor())
model = classifier.fit(trainingDateset)
result = model.transform(testingDataset)
unit_test_utils.assert_data_frames_are_identical(referenceDataset, result)
def testH2OGAMClassifierBehavesTheSameAsGenericH2OGAMOnStringLabelColumn(
prostateDataset):
[trainingDateset, testingDataset] = prostateDataset.randomSplit([0.9, 0.1],
42)
gam = setParamtersForProblemSpecificTests(H2OGAM())
referenceModel = gam.fit(
trainingDateset.withColumn("CAPSULE",
col("CAPSULE").cast("string")))
referenceDataset = referenceModel.transform(testingDataset)
classifier = setParamtersForProblemSpecificTests(H2OGAMClassifier())
model = classifier.fit(trainingDateset)
result = model.transform(testingDataset)
unit_test_utils.assert_data_frames_are_identical(referenceDataset, result)
def testPipelineSerialization(prostateDataset):
algo = H2OGAM(featuresCols=["DPROS", "DCAPS", "RACE", "GLEASON"],
gamCols=["PSA", "AGE"],
labelCol="CAPSULE",
seed=1,
splitRatio=0.8)
pipeline = Pipeline(stages=[algo])
pipeline.write().overwrite().save("file://" +
os.path.abspath("build/gam_pipeline"))
loadedPipeline = Pipeline.load("file://" +
os.path.abspath("build/gam_pipeline"))
model = loadedPipeline.fit(prostateDataset)
model.write().overwrite().save("file://" +
os.path.abspath("build/gam_pipeline_model"))
loadedModel = PipelineModel.load(
"file://" + os.path.abspath("build/gam_pipeline_model"))
loadedModel.transform(prostateDataset).count()
def testPipelineSerializationGAM(prostateDataset):
gridSearchTester(H2OGAM().setLabelCol("AGE").setGamCols(["PSA"]),
prostateDataset)
def createInitialGamDefinition():
return H2OGAM(featuresCols=featuresCols,
labelCol="CAPSULE",
seed=1,
splitRatio=0.8,
gamCols=["PSA", "AGE"])