def import_svmlight(path, headers=""):
raw = h2o.lazy_import(path)
if settings.debug and len(headers) < 100:
print utils.time() + "import with headers: " + str(headers)
#parsesetup = h2o.parse_setup(raw,column_names=headers)
parsesetup = h2o.parse_setup(
raw
) # Issue: H2O 3.8 tests length of header vs. columns, but still imports the "pseudotarget" additionally
parsesetup['parse_type'] = 'SVMLight'
loaded_frame = h2o.parse_raw(parsesetup)
if settings.debug:
print "......HEader length: " + str(len(headers))
print "......Frame imported: " + str(loaded_frame.ncol)
if (len(headers) > loaded_frame.ncol):
n = len(headers) - loaded_frame.ncol
print "Remove last " + str(n) + " header entries"
del headers[-n:]
loaded_frame.set_names(headers) #Workaround, Set names now
print "First column: " + loaded_frame.names[
0] #needed because lazy name setting
if settings.debug and len(headers) < 100: loaded_frame.head(show=True)
loaded_frame.pop(0) #remove first ('pseudotarget') columnn
#if loaded_frame.ncol>len(headers)-1: #workaround: H2O reads info from svmlight into columns -> remove everything that is not in headers
# delete = []
# for i in xrange(len(headers)-1,loaded_frame.ncol):
# delete.append(loaded_frame.names[i])
# loaded_frame = remove_vecs(loaded_frame,delete)
if settings.debug and len(headers) < 100: loaded_frame.head(show=True)
return loaded_frame
def h2olazy_import():
"""
Python API test: h2o.lazy_import(path)
"""
training_data = h2o.lazy_import(
pyunit_utils.locate("smalldata/prostate/prostate_cat.csv"))
assert_is_type(training_data, list)
def __init__(self, mojo_path=None):
"""
Create a new H2OMojoPipeline object.
:param mojo_path path to a MOJO file.
"""
assert_is_type(mojo_path, str)
self.pipeline_id = h2o.lazy_import(mojo_path)
def h2olazy_import():
"""
Python API test: h2o.lazy_import(path)
"""
try:
training_data = h2o.lazy_import(
pyunit_utils.locate("smalldata/prostate/prostate_cat.csv"))
except Exception as e:
assert False, "h2o.lazy_import() command is not working."
def continuous_or_categorical():
fraw = h2o.lazy_import(tests.locate("smalldata/jira/hexdev_29.csv"))
fsetup = h2o.parse_setup(fraw)
fsetup["column_types"][0] = "ENUM"
fsetup["column_types"][1] = "ENUM"
fsetup["column_types"][2] = "ENUM"
df_hex = h2o.parse_raw(fsetup)
df_hex.summary()
assert (df_hex['h1'].isfactor())
assert (df_hex['h2'].isfactor())
assert (df_hex['h3'].isfactor())
def from_file(file=str):
"""
Creates new Generic model by loading existing embedded model into library, e.g. from H2O MOJO.
The imported model must be supported by H2O.
:param file: A string containing path to the file to create the model from
:return: H2OGenericEstimator instance representing the generic model
"""
from h2o import lazy_import, get_frame
model_key = lazy_import(file)
model_bytes_frame = get_frame(model_key[0])
model = H2OGenericEstimator(model_key=model_bytes_frame)
model.train()
return model
def hexdev_394():
path = tests.locate("smalldata/covtype/covtype.20k.data")
trainraw = h2o.lazy_import(path)
tsetup = h2o.parse_setup(trainraw)
tsetup["column_types"][10] = "ENUM"
tsetup["column_types"][11] = "ENUM"
tsetup["column_types"][12] = "ENUM"
train = h2o.parse_raw(tsetup)
cols = train.col_names # This returned space for first column name
x_cols = [colname for colname in cols if colname != "C55"]
x_cols
splits = train.split_frame()
newtrain = splits[0]
newvalid = splits[1]
newtrain_x = newtrain[x_cols]
newtrain_y = newtrain[54].asfactor()
newvalid_x = newvalid[x_cols]
newvalid_y = newvalid[54].asfactor()
my_gbm = h2o.gbm(y=newtrain_y,
validation_y=newvalid_y,
x=newtrain_x,
validation_x=newvalid_x,
distribution = "multinomial",
ntrees=100,
learn_rate=0.1,
max_depth=6)
split1, split2 = train.split_frame()
newtrain_x = split1[x_cols]
newtrain_y = split1[54].asfactor()
newvalid_x = split2[x_cols]
newvalid_y = split2[54].asfactor()
my_gbm = h2o.gbm(y=newtrain_y,
validation_y=newvalid_y,
x=newtrain_x,
validation_x=newvalid_x,
distribution = "multinomial",
ntrees=100,
learn_rate=0.1,
max_depth=6)
print "KEEPING FRAME???"
print train._keep
def continuous_or_categorical():
fraw = h2o.lazy_import(h2o.locate("smalldata/jira/hexdev_29.csv"))
fsetup = h2o.parse_setup(fraw)
fsetup["column_types"][0] = "ENUM"
fsetup["column_types"][1] = "ENUM"
fsetup["column_types"][2] = "ENUM"
df_hex = h2o.parse_raw(fsetup)
df_hex.summary()
assert (df_hex['h1'].isfactor())
assert (df_hex['h2'].isfactor())
assert (df_hex['h3'].isfactor())
def read_csv(file_path, destination_frame, header=(-1,0,1), separator="", column_names=None, column_types=None, na_strings=None):
"""
Build an H2OFrame from parsing a CSV at file_path. This path is relative to
the H2O cluster, NOT the local Python process
:param file_path: A remote path to a data source. Data is cluster-local.
:param destination_frame: The result *Key* name in the H2O cluster
"""
rawkey = h2o.lazy_import(file_path)
res = H2OFrame._parse(rawkey, destination_frame, header, separator, column_names, column_types, na_strings)
nrows = res.nrow
ncols = res.ncol
if isinstance(file_path, str): print "Imported {}. Parsed {} rows and {} cols".format(file_path,"{:,}".format(nrows), "{:,}".format(ncols))
else: h2o.H2ODisplay([["File"+str(i+1),f] for i,f in enumerate(file_path)],None, "Parsed {} rows and {} cols".format("{:,}".format(nrows), "{:,}".format(ncols)))
return res
def test(x, y, output_test, strip_part, algo_name, generic_algo_name):
airlines = h2o.import_file(
path=pyunit_utils.locate("smalldata/testng/airlines_train.csv"))
gbm = H2ODeepLearningEstimator(epochs=1)
gbm.train(x=x, y=y, training_frame=airlines, validation_frame=airlines)
print(gbm)
with Capturing() as original_output:
gbm.show()
original_model_filename = tempfile.mkdtemp()
original_model_filename = gbm.download_mojo(original_model_filename)
key = h2o.lazy_import(original_model_filename)
fr = h2o.get_frame(key[0])
generic_mojo_model = H2OGenericEstimator(model_key=fr)
generic_mojo_model.train()
compare_params(gbm, generic_mojo_model)
print(generic_mojo_model)
with Capturing() as generic_output:
generic_mojo_model.show()
output_test(str(original_output), str(generic_output), strip_part,
algo_name, generic_algo_name)
predictions = generic_mojo_model.predict(airlines)
assert predictions is not None
assert predictions.nrows == 24421
assert generic_mojo_model._model_json["output"][
"model_summary"] is not None
assert len(generic_mojo_model._model_json["output"]
["model_summary"]._cell_values) > 0
# Test constructor generating the model from existing MOJO file
generic_mojo_model_from_file = H2OGenericEstimator.from_file(
original_model_filename)
assert generic_mojo_model_from_file is not None
predictions = generic_mojo_model_from_file.predict(airlines)
assert predictions is not None
assert predictions.nrows == 24421
assert generic_mojo_model_from_file._model_json["output"][
"model_summary"] is not None
assert len(generic_mojo_model_from_file._model_json["output"]
["model_summary"]._cell_values) > 0
generic_mojo_filename = tempfile.mkdtemp("zip", "genericMojo")
generic_mojo_filename = generic_mojo_model_from_file.download_mojo(
path=generic_mojo_filename)
assert os.path.getsize(generic_mojo_filename) == os.path.getsize(
original_model_filename)
def stackedensemble_mojo_model_test():
train = h2o.import_file(
pyunit_utils.locate("smalldata/iris/iris_train.csv"))
test = h2o.import_file(pyunit_utils.locate("smalldata/iris/iris_test.csv"))
x = train.columns
y = "species"
nfolds = 2
gbm = H2OGradientBoostingEstimator(nfolds=nfolds,
fold_assignment="Modulo",
keep_cross_validation_predictions=True)
gbm.train(x=x, y=y, training_frame=train)
rf = H2ORandomForestEstimator(nfolds=nfolds,
fold_assignment="Modulo",
keep_cross_validation_predictions=True)
rf.train(x=x, y=y, training_frame=train)
se = H2OStackedEnsembleEstimator(training_frame=train,
validation_frame=test,
base_models=[gbm.model_id, rf.model_id])
se.train(x=x, y=y, training_frame=train)
print(se)
with Capturing() as original_output:
se.show()
original_model_filename = tempfile.mkdtemp()
original_model_filename = se.download_mojo(original_model_filename)
key = h2o.lazy_import(original_model_filename)
fr = h2o.get_frame(key[0])
generic_mojo_model = H2OGenericEstimator(model_key=fr)
generic_mojo_model.train()
compare_params(se, generic_mojo_model)
predictions = generic_mojo_model.predict(test)
assert predictions is not None
# Test constructor generating the model from existing MOJO file
generic_mojo_model_from_file = H2OGenericEstimator.from_file(
original_model_filename)
assert generic_mojo_model_from_file is not None
predictions = generic_mojo_model_from_file.predict(test)
assert predictions is not None
generic_mojo_filename = tempfile.mkdtemp("zip", "genericMojo")
generic_mojo_filename = generic_mojo_model_from_file.download_mojo(
path=generic_mojo_filename)
assert os.path.getsize(generic_mojo_filename) == os.path.getsize(
original_model_filename)
def test(x, y, output_test, strip_part, algo_name, generic_algo_name):
airlines = h2o.import_file(
path=pyunit_utils.locate("smalldata/testng/airlines_train.csv"))
xgb = H2OXGBoostEstimator(ntrees=1, nfolds=3)
xgb.train(x=x, y=y, training_frame=airlines, validation_frame=airlines)
print(xgb)
with Capturing() as original_output:
xgb.show()
original_model_filename = tempfile.mkdtemp()
original_model_filename = xgb.download_mojo(original_model_filename)
key = h2o.lazy_import(original_model_filename)
fr = h2o.get_frame(key[0])
model = H2OGenericEstimator(model_key=fr)
model.train()
print(model)
with Capturing() as generic_output:
model.show()
output_test(str(original_output), str(generic_output), strip_part,
algo_name, generic_algo_name)
predictions = model.predict(airlines)
assert predictions is not None
assert predictions.nrows == 24421
assert model._model_json["output"]["variable_importances"] is not None
assert len(
model._model_json["output"]["variable_importances"]._cell_values) > 0
assert model._model_json["output"]["model_summary"] is not None
assert len(model._model_json["output"]["model_summary"]._cell_values) > 0
# Test constructor generating the model from existing MOJO file
model = H2OGenericEstimator.from_file(original_model_filename)
assert model is not None
predictions = model.predict(airlines)
assert predictions is not None
assert predictions.nrows == 24421
assert model._model_json["output"]["variable_importances"] is not None
assert len(
model._model_json["output"]["variable_importances"]._cell_values) > 0
assert model._model_json["output"]["model_summary"] is not None
assert len(model._model_json["output"]["model_summary"]._cell_values) > 0
generic_mojo_filename = tempfile.mkdtemp("zip", "genericMojo")
generic_mojo_filename = model.download_mojo(path=generic_mojo_filename)
assert os.path.getsize(generic_mojo_filename) == os.path.getsize(
original_model_filename)
def hexdev_394():
path = tests.locate("smalldata/covtype/covtype.20k.data")
trainraw = h2o.lazy_import(path)
tsetup = h2o.parse_setup(trainraw)
tsetup["column_types"][10] = "ENUM"
tsetup["column_types"][11] = "ENUM"
tsetup["column_types"][12] = "ENUM"
train = h2o.parse_raw(tsetup)
cols = train.col_names # This returned space for first column name
x_cols = [colname for colname in cols if colname != "C55"]
x_cols
splits = train.split_frame()
newtrain = splits[0]
newvalid = splits[1]
newtrain_x = newtrain[x_cols]
newtrain_y = newtrain[54].asfactor()
newvalid_x = newvalid[x_cols]
newvalid_y = newvalid[54].asfactor()
my_gbm = h2o.gbm(y=newtrain_y,
validation_y=newvalid_y,
x=newtrain_x,
validation_x=newvalid_x,
distribution="multinomial",
ntrees=100,
learn_rate=0.1,
max_depth=6)
split1, split2 = train.split_frame()
newtrain_x = split1[x_cols]
newtrain_y = split1[54].asfactor()
newvalid_x = split2[x_cols]
newvalid_y = split2[54].asfactor()
my_gbm = h2o.gbm(y=newtrain_y,
validation_y=newvalid_y,
x=newtrain_x,
validation_x=newvalid_x,
distribution="multinomial",
ntrees=100,
learn_rate=0.1,
max_depth=6)
print "KEEPING FRAME???"
print train._keep
def mojo_model_test():
# GBM
airlines = h2o.import_file(
path=pyunit_utils.locate("smalldata/testng/airlines_train.csv"))
gbm = H2OGradientBoostingEstimator(ntrees=1)
gbm.train(x=["Origin", "Dest"], y="IsDepDelayed", training_frame=airlines)
original_model_filename = tempfile.mkdtemp()
original_model_filename = gbm.download_mojo(original_model_filename)
key = h2o.lazy_import(original_model_filename)
fr = h2o.get_frame(key[0])
model = H2OGenericEstimator(model_key=fr)
model.train()
predictions = model.predict(airlines)
assert predictions is not None
assert predictions.nrows == 24421
assert model._model_json["output"]["variable_importances"] is not None
assert len(
model._model_json["output"]["variable_importances"]._cell_values) > 0
assert model._model_json["output"]["model_summary"] is not None
assert len(model._model_json["output"]["model_summary"]._cell_values) > 0
# Test constructor generating the model from existing MOJO file
model = H2OGenericEstimator.from_file(original_model_filename)
assert model is not None
predictions = model.predict(airlines)
assert predictions is not None
assert predictions.nrows == 24421
assert model._model_json["output"]["variable_importances"] is not None
assert len(
model._model_json["output"]["variable_importances"]._cell_values) > 0
assert model._model_json["output"]["model_summary"] is not None
assert len(model._model_json["output"]["model_summary"]._cell_values) > 0
generic_mojo_filename = tempfile.mkdtemp("zip", "genericMojo")
generic_mojo_filename = model.download_mojo(path=generic_mojo_filename)
assert os.path.getsize(generic_mojo_filename) == os.path.getsize(
original_model_filename)
def mojo_model_test():
# GBM
airlines = h2o.import_file(path=pyunit_utils.locate("smalldata/testng/airlines_train.csv"))
gbm = H2OGradientBoostingEstimator(ntrees = 1)
gbm.train(x = ["Origin", "Dest"], y = "IsDepDelayed", training_frame=airlines)
original_model_filename = tempfile.mkdtemp()
original_model_filename = gbm.download_mojo(original_model_filename)
key = h2o.lazy_import(original_model_filename)
fr = h2o.get_frame(key[0])
model = H2OGenericEstimator(model_key = fr)
model.train()
predictions = model.predict(airlines)
assert predictions is not None
assert predictions.nrows == 24421
assert model._model_json["output"]["variable_importances"] is not None
assert len(model._model_json["output"]["variable_importances"]._cell_values) > 0
assert model._model_json["output"]["model_summary"] is not None
assert len(model._model_json["output"]["model_summary"]._cell_values) > 0
# Test constructor generating the model from existing MOJO file
model = H2OGenericEstimator.from_file(original_model_filename)
assert model is not None
predictions = model.predict(airlines)
assert predictions is not None
assert predictions.nrows == 24421
assert model._model_json["output"]["variable_importances"] is not None
assert len(model._model_json["output"]["variable_importances"]._cell_values) > 0
assert model._model_json["output"]["model_summary"] is not None
assert len(model._model_json["output"]["model_summary"]._cell_values) > 0
generic_mojo_filename = tempfile.mkdtemp("zip", "genericMojo");
generic_mojo_filename = model.download_mojo(path=generic_mojo_filename)
assert os.path.getsize(generic_mojo_filename) == os.path.getsize(original_model_filename)
def _import_parse(path, destination_frame, header, sep, column_names, column_types, na_strings): rawkey = h2o.lazy_import(path) return H2OFrame._parse(rawkey,destination_frame, header, sep, column_names, column_types, na_strings)
def h2olazy_import():
"""
Python API test: h2o.lazy_import(path)
"""
training_data = h2o.lazy_import(pyunit_utils.locate("smalldata/prostate/prostate_cat.csv"))
assert_is_type(training_data, list)