def test_glm(cas_session, boston_dataset):
target = {
'tool': 'SAS Visual Analytics',
'targetVariable': 'Price',
'scoreCodeType': 'ds2MultiType',
'function': 'prediction',
'algorithm': 'Linear regression'
}
cas_session.loadactionset('regression')
cas_session.loadactionset('astore')
tbl = cas_session.upload(boston_dataset).casTable
tbl.regression.glm(target='Price',
inputs=list(boston_dataset.columns[:-1]),
savestate='astore')
desc = cas_session.astore.describe(rstore='astore', epcode=True)
props = _get_model_properties(desc)
# Verify properties are set
for k, v in six.iteritems(target):
assert props[k] == v
files = create_files_from_astore(cas_session.CASTable('astore'))
check_input_variables(files, BOSTON_INPUT_VARS)
def test_logistic(cas_session, iris_dataset):
target = {
'tool': 'SAS Visual Analytics',
'targetVariable': 'Species',
'scoreCodeType': 'ds2MultiType',
'function': 'classification',
'algorithm': 'Logistic regression'
}
cas_session.loadactionset('regression')
cas_session.loadactionset('astore')
tbl = cas_session.upload(iris_dataset).casTable
tbl.regression.logistic(target='Species',
inputs=list(iris_dataset.columns[:-1]),
savestate='astore')
desc = cas_session.astore.describe(rstore='astore', epcode=True)
props = _get_model_properties(desc)
for k, v in six.iteritems(target):
assert props[k] == v
files = create_files_from_astore(cas_session.CASTable('astore'))
check_input_variables(files, IRIS_INPUT_VARS)
def test_bayesnet_classification(cas_session, iris_dataset):
target = {
'tool': 'SAS Visual Data Mining and Machine Learning',
'targetVariable': 'Species',
'scoreCodeType': 'ds2MultiType',
'function': 'classification',
'algorithm': 'Bayesian network'
}
cas_session.loadactionset('bayesianNetClassifier')
cas_session.loadactionset('astore')
tbl = cas_session.upload(iris_dataset).casTable
tbl.bayesianNetClassifier.bnet(target='Species',
inputs=list(iris_dataset.columns[:-1]),
savestate='astore')
desc = cas_session.astore.describe(rstore='astore', epcode=True)
props = _get_model_properties(desc)
for k, v in six.iteritems(target):
assert props[k] == v
files = create_files_from_astore(cas_session.CASTable('astore'))
check_input_variables(files, IRIS_INPUT_VARS)
def test_svm_regression(cas_session, boston_dataset):
target = {
'tool': 'SAS Visual Data Mining and Machine Learning',
'targetVariable': 'Price',
'scoreCodeType': 'ds2MultiType',
'function': 'prediction',
'algorithm': 'Support vector machine'
}
cas_session.loadactionset('svm')
cas_session.loadactionset('astore')
tbl = cas_session.upload(boston_dataset).casTable
tbl.svm.svmTrain(target='Price',
inputs=list(boston_dataset.columns[:-1]),
saveState='astore')
desc = cas_session.astore.describe(rstore='astore', epcode=True)
props = _get_model_properties(desc)
for k, v in six.iteritems(target):
assert props[k] == v
files = create_files_from_astore(cas_session.CASTable('astore'))
check_input_variables(files, BOSTON_INPUT_VARS)
def test_neuralnet_regression(cas_session, boston_dataset):
target = {
'tool': 'SAS Visual Data Mining and Machine Learning',
'targetVariable': 'Price',
'scoreCodeType': 'ds2MultiType',
'function': 'prediction',
'algorithm': 'Neural network'
}
cas_session.loadactionset('neuralnet')
cas_session.loadactionset('astore')
tbl = cas_session.upload(boston_dataset).casTable
tbl.neuralNet.annTrain(target='Price',
inputs=list(boston_dataset.columns[:-1]),
# modelTable='network',
arch='MLP',
hiddens=[2],
combs=['linear'],
casout='network')
# savestate='astore')
pytest.skip('Implement. How to get an astore?')
desc = cas_session.astore.describe(rstore='astore', epcode=True)
props = _get_model_properties(desc)
for k, v in six.iteritems(target):
assert props[k] == v
files = create_files_from_astore(cas_session.CASTable('astore'))
check_input_variables(files, BOSTON_INPUT_VARS)
def test_gradboost_binary_classification(cas_session, cancer_dataset):
target = {
'tool': 'SAS Visual Data Mining and Machine Learning',
'targetVariable': 'Type',
'scoreCodeType': 'ds2MultiType',
'function': 'classification',
'algorithm': 'Gradient boosting',
'targetLevel': 'binary'
}
cas_session.loadactionset('decisiontree')
cas_session.loadactionset('astore')
tbl = cas_session.upload(cancer_dataset).casTable
tbl.decisiontree.gbtreetrain(target='Type',
inputs=list(cancer_dataset.columns[:-1]),
savestate='astore')
desc = cas_session.astore.describe(rstore='astore', epcode=True)
props = _get_model_properties(desc)
for k, v in six.iteritems(target):
assert props[k] == v
files = create_files_from_astore(cas_session.CASTable('astore'))
check_input_variables(files, CANCER_INPUT_VARS)
def test_dtree_regression(cas_session, boston_dataset):
target = {
'tool': 'SAS Visual Analytics',
'targetVariable': 'Price',
'scoreCodeType': 'ds2MultiType',
'function': 'prediction',
'algorithm': 'Decision tree'
}
cas_session.loadactionset('decisiontree')
cas_session.loadactionset('astore')
tbl = cas_session.upload(boston_dataset).casTable
tbl.decisiontree.dtreetrain(target='Price',
inputs=list(boston_dataset.columns[:-1]),
casout='tree')
pytest.skip('Implement. How to get an astore?')
cas_session.decisiontree.dtreeExportModel(modelTable='tree',
casout='astore')
desc = cas_session.astore.describe(rstore='astore', epcode=True)
props = _get_model_properties(desc)
for k, v in six.iteritems(target):
assert props[k] == v
files = create_files_from_astore(cas_session.CASTable('astore'))
check_input_variables(files, BOSTON_INPUT_VARS)