def rapidminer_quick_training(go_url, gouser, gopassword, input_data, label,
selection_criteria, max_min_crietria_selector,
platform):
from rapidminer_go_python import rapidminergoclient as amw
# To get the AMW instance
client = amw.RapidMinerGoClient(go_url, gouser, gopassword)
data = client.convert_json_to_dataframe(input_data)
trainingResult = client.quick_automodel(data, label, AUTODEPLOY,
selection_criteria,
max_min_crietria_selector)
if platform == 'tabprep':
return trainingResult
prediction = []
# Number of records in input
max_data_length = len(data.index)
# Adding survived to a list
for i in range(0, max_data_length):
prediction.append(data.iloc[i][label])
print('returning result')
return prediction
def rapidminer_train(go_url, gouser, gopassword, input_data, label,
cost_matrix, high_value, low_value, selection_criteria,
max_min_crietria_selector, platform):
from rapidminer_go_python import rapidminergoclient as amw
LABEL_ATTRIBUTE = label
client = amw.RapidMinerGoClient(go_url, gouser, gopassword)
data = client.convert_json_to_dataframe(input_data)
dataId = client.add_dataFrame(data)[DATA_ID]
modelingTaskID = client.create_modeling_task(dataId)[DATA_ID]
# setting label
client.set_label(modelingTaskID, LABEL_ATTRIBUTE)
client.set_class_interest(modelingTaskID, high_value, low_value)
client.set_cost_matrix(modelingTaskID, cost_matrix)
print('TModeling askID:' + modelingTaskID)
# Initiating model training
client.start_execution(modelingTaskID)
print('ExecutingModel...')
# Obtaining the trained model results
client.get_execution_result(modelingTaskID)
# To find the best model**add or remove features if needed to get a value of more or less deep rooted in Json*
bestModel = client.determine_best_model(selection_criteria,
max_min_crietria_selector)
# Deploying the best model
global depID
depID = client.deploy_model(modelingTaskID, bestModel)
status = 'Failed'
if str(depID) != '':
status = 'Success'
url_result = client.SERVER + '/am/modeling/' + str(
modelingTaskID) + '/results'
# Binding DeploymentID, Status and Best Model together in a dictionary to return as a output
out_result = {
MODELING_ID: modelingTaskID,
DEPLOYMENT_ID: depID,
STATUS: status,
MODEL: bestModel,
URL: url_result
}
client.convert_json_to_dataframe(out_result)
print('DeploymentID:' + str(depID))
if platform == 'tabprep':
return out_result
prediction = []
# Number of records in input
max_data_length = len(data.index)
# Adding survived to a list
for i in range(0, max_data_length):
prediction.append(data.iloc[i][label])
print('returning result')
return prediction
def rapidminer_score(go_url, gouser, gopassword, inputScoreData, label, depID):
print('Inside Score Method, DeploymentID ' + depID)
from rapidminer_go_python import rapidminergoclient as amw
global client
client = amw.RapidMinerGoClient(go_url, gouser, gopassword)
PREDICTION = 'prediction(' + label + ')'
# passing the test data to deployed model to score
scoreResult = client.score(inputScoreData, depID)
# converting result json to dataframe
result = client.convert_json_to_dataframe(scoreResult['data'])
req = client.convert_json_to_dataframe(inputScoreData)
# List to add the result data
prediction = []
# Number of records in input
max_length = len(req.index)
# Adding confidence and predictions to a list
for i in range(0, max_length):
prediction.append(result.iloc[i][PREDICTION])
print('Scoring Completed Successfully')
return prediction