def main(model_path):
h2o.init(nthreads=-1, max_mem_size=16)
serve(model_path)
from h2o import h2o
h2o.init()
# example adapted from http://h2o-release.s3.amazonaws.com/h2o-dev/rel-shannon/2/docs-website/h2o-py/docs/h2o.html#models
fr = h2o.import_file(
path=
"https://raw.githubusercontent.com/h2oai/h2o-2/master/smalldata/logreg/prostate.csv"
)
r = fr[0].runif()
train = fr[r < 0.70]
test = fr[r >= 0.70]
train["CAPSULE"] = train["CAPSULE"].asfactor()
test["CAPSULE"] = test["CAPSULE"].asfactor()
m = h2o.H2OGeneralizedLinearEstimator(family='binomial', alpha=[0.5])
m.train(x=["AGE", "RACE", "PSA", "VOL", "GLEASON"],
y="CAPSULE",
training_frame=train)
m.show()
def __init__(self, ip, port):
self.ip = ip
self.port = port
# initialize the http connection, needed for the communication via REST endpoints
h2o.init(ip=ip,port=port)
self.scala_session_id = H2OContext.init_scala_int_session()
import pandas as pd
from h2o import h2o
# Removing existing data from H2O Cluster
h2o.init(ip="localhost", port=54321)
h2o.remove_all()
# Loading HR Analytics Data from CSV File
full_data_frame = h2o.H2OFrame(
pd.read_csv("dataset/HR_comma_sep.csv", index_col=None, header=0))
# Defining categorical features
feature_columns = [
'left', 'Work_accident', 'promotion_last_5years', 'department'
]
# Defining continuous features
continuous_feature_columns = [
'satisfaction_level', 'last_evaluation', 'number_project',
'average_montly_hours', 'time_spend_company', 'salary'
]
training_data_frame, test_data_frame = full_data_frame.split_frame(ratios=[.8])
training_data_frame[feature_columns] = training_data_frame[
feature_columns].asfactor()
test_data_frame[feature_columns] = test_data_frame[feature_columns].asfactor()
print(training_data_frame[0, :])
print(test_data_frame[0, :])
def __init__(self, ip, port, initialize = True):
self.ip = ip
self.port = port
# initialize the http connection, needed for the communication via REST endpoints
if initialize:
h2o.init(ip=ip, port=port)
