def build_scenario(build_plan,
filter_on='ready',
connection='dataiku_workspace',
ref_table='referentialclient',
ref_project='DIReferential',
add_ecrm_context=True,
finish_on_client=None,
single_client=None):
scenario = Scenario()
if not isinstance(filter_on, list):
filter_on = [filter_on]
project_name = dataiku.default_project_key()
project_key = dataiku.api_client().get_project(project_name)
local_variables = project_key.get_variables()['local']
env = local_variables['env']
kut.display_message('reading client context referential')
executor = SQLExecutor2(connection=connection)
sql_query_referential_client = "SELECT * FROM " + '_'.join(
[env, ref_project, ref_table])
client_ref = executor.query_to_df(sql_query_referential_client)
filter_query = ' & '.join(filter_on)
client_ref = client_ref.query(filter_query) if filter_query else client_ref
kut.display_message('Client ready for automation : ' +
client_ref.clientName.unique())
kut.display_message('run configuration')
print(build_plan)
if not pd.isnull(finish_on_client):
finish_client = client_ref[client_ref.clientName == finish_on_client]
if len(finish_client) == 0:
kut.display_message(
'finish client not found in plan ' + finish_on_client +
' is the client name valid ?'
) # Example: load a DSS dataset as a Pandas dataframe
other_clients = client_ref[client_ref.clientName != finish_on_client]
client_ref = pd.concat([other_clients, finish_client],
ignore_index=True)
success = []
if single_client is not None:
requested_client = client_ref[client_ref.clientName == single_client]
if not len(single_client):
kut.display_message(
'requested single client is not found,building all allowed clients'
)
else:
client_ref = requested_client
for index, client_row in client_ref.iterrows():
variables = set_client_context(client_row=client_row,
add_ecrm_context=add_ecrm_context,
connection=connection)
client_name = variables['local']['clientName']
kut.display_message('starting builds on ' + client_name)
run_scenario(table_plan=build_plan, scenario=scenario)
success.append(client_name)
scenario.set_global_variables(successfullRun=success)
print('done_________________' + client_name)
return success
from dataiku.scenario import Scenario scenario = Scenario() trigger_type = scenario.get_trigger_type() trigger_name = scenario.get_trigger_name() # depending on the trigger type, different metadata can be available trigger_params = scenario.get_trigger_params()
# this part can be used in a custom scenario's script, or in a "Execute python" step in a step-based scenario
import dataiku
from dataiku.scenario import Scenario
s = Scenario()
dataset_name = 'input_partitioned'
output_name = 'output'
# fetch the partitions
ds = dataiku.Dataset(dataset_name)
all_partitions = ds.list_partitions()
print("Dataset %s has %s partitions" % (dataset_name, len(all_partitions)))
# maybe filter partitions, depending on your usage
partitions_to_build = all_partitions
# build the variable's value as a comma separated string
partition_list_value = ','.join(partitions_to_build)
s.set_scenario_variables(partition_list=partition_list_value)
# in a step-based scenario:
# add a build step to build the output dataset, and set ${partition_list} as the partition identifier
# in a custom scenario:
# launch the build
s.build_dataset(output_name, partitions='${partition_list}')
# alternatively, in a custom scenario, you can pass the value directly, without using a variable
#s.build_dataset(output_name, partitions=partition_list_value)
from dataiku.scenario import Scenario scenario = Scenario() # Note that you must be admin to update global variables scenario.set_global_variables(var1="val1", var2=3)
from dataiku.scenario import Scenario scenario = Scenario() # Note that you must be admin to update global variables scenario.run_global_variables_update()
from dataiku.scenario import Scenario
scenario = Scenario()
# Partitions are specified using the partitions spec syntax
scenario.build_dataset("mydataset", partitions="partition1|partition2")
from dataiku.scenario import Scenario
import time
scenario = Scenario()
step1 = scenario.build_dataset("mydataset1", async=True)
step2 = scenario.build_dataset("mydataset2", async=True)
while not step1.is_done() or not step2.is_done():
# do something while waiting
time.sleep(1)
from dataiku.scenario import Scenario
scenario = Scenario()
# The id of the model is visible in the URL of the model
scenario.train_model("mymodelid")
# This sample code helps you get started with the custom scenario API.
#For more details and samples, please see our Documentation
from dataiku.scenario import Scenario
# The Scenario object is the main handle from which you initiate steps
scenario = Scenario()
# A few example steps follow
# Building a dataset
scenario.build_dataset("customers_prepared", partitions="2015-01-03")
# Controlling the train of a dataset
train_ret = scenario.train_model("uSEkldfsm")
trained_model = train_ret.get_trained_model()
performance = trained_model.get_new_version_metrics().get_performance_values()
if performance["AUC"] > 0.85:
trained_model.activate_new_version()
# Sending custom reports
sender = scenario.get_message_sender("mail-scenario", "local-mail") # A messaging channel
sender.set_params(sender="*****@*****.**", recipient="*****@*****.**")
sender.send(subject="The scenario is doing well", message="All is good")
###########################################################################################
# !! CUSTOM SCENARIO EXAMPLE !! #
# See https://doc.dataiku.com/dss/latest/scenarios/custom_scenarios.html for more details #
###########################################################################################
import time
import dataiku
from dataiku.scenario import Scenario, BuildFlowItemsStepDefHelper
from dataikuapi.dss.future import DSSFuture
TIMEOUT_SECONDS = 3600
s = Scenario()
# Replace this commented block by your Scenario steps
# Example: build a Dataset
step_handle = s.build_dataset("your_dataset_name", asynchronous=True)
start = time.time()
while not step_handle.is_done():
end = time.time()
print("Duration: {}s".format(end - start))
if end - start > TIMEOUT_SECONDS:
f = DSSFuture(dataiku.api_client(), step_handle.future_id)
f.abort()
raise Exception("Scenario was aborted because it took too much time.")
from dataiku.scenario import Scenario
scenario = Scenario()
scenario.run_dataset_checks("mydataset")
import dataiku
import pandas as pd, numpy as np
from dataiku import pandasutils as pdu
import uuid
from dataiku.scenario import Scenario
PARTNER = 'IHG'
MODEL_TYPE = 'BG'
proj_handle = dataiku.api_client().get_project("IHG_AUTO_V1_1")
proj_var = proj_handle.get_variables()
# Create the main handle to interact with the scenario
scen = Scenario()
step = scen.get_previous_steps_outputs()
# Score #
result_date = ([d for d in step if d['stepName'] == 'query_date'])[0]['result']
new_score_date = result_date['rows'][0][0]
# train #
result_train_ref_date = ([
d for d in step if d['stepName'] == 'train_ref_date'
])[0]['result']
new_train_date = result_train_ref_date['rows'][0][0]
# valid #
result_valid_ref_date = ([
d for d in step if d['stepName'] == 'valid_ref_date'
])[0]['result']
new_valid_date = result_valid_ref_date['rows'][0][0]
#previous value
cur_score_date = proj_var["standard"]["scoring_file_date"]
cur_train_date = proj_var["standard"]["train_file_date"]
# This sample code helps you get started with the custom scenario API.
#For more details and samples, please see our Documentation
from dataiku.scenario import Scenario
# The Scenario object is the main handle from which you initiate steps
scenario = Scenario()
# A few example steps follow
# Building a dataset
scenario.build_dataset("scores",
build_mode="RECURSIVE_FORCED_BUILD",
project_key="FRAUD_MODEL")
scenario.build_dataset("unseen_scored", build_mode="RECURSIVE_FORCED_BUILD")
from dataiku.scenario import Scenario
scenario = Scenario()
if scenario.get_trigger_type() == 'exec_sql':
trigger_params = scenario.get_trigger_params()
# the list of the columns in the query output
columns = trigger_params['result']['rows']
# columns contain name and type
print("\t".join([column['name'] for column in columns]))
print("\t".join([column['type'] for column in columns]))
# the rows in the result, as an array of array of strings
rows = trigger_params['result']['rows']
for row in rows:
print("\t".join(row))
from dataiku.scenario import Scenario
scenario = Scenario()
# The id of the folder is visible in the URL of the managed folder
scenario.build_folder("myfolderid")
# this part can be used in a custom scenario's script, or in a "Execute python" step in a step-based scenario
import dataiku
from dataiku.scenario import Scenario
s = Scenario()
#SET DATASET NAMES
# The dataset that gets its partitions added outside of dataiku, or through
# another process should be set in the updated variable
###
# The dataset that the original dataset writes to should be the old variable
updated = 'customers'
old = 'customers_prepared'
# fetch the partitions
def partition_list(dataset):
return dataiku.Dataset(dataset).list_partitions()
# Get partitions in new set that aren't in old set
partitions_to_build = list(
set(partition_list(updated)) - set(partition_list(old)))
# build the variable's value as a comma separated string
partition_list_value = ','.join(partitions_to_build)
s.set_scenario_variables(partition_list=partition_list_value)
print partition_list_value
# in a step-based scenario:
from dataiku.scenario import Scenario
scenario = Scenario()
scenario.execute_sql("connection", "UPDATE TABLE t SET ...")
# Get senario handler
from dataiku.scenario import Scenario
scenario = Scenario()
# Create a message sender
sender = scenario.get_message_sender(channel_id="gmail") # A messaging channel
# Define your attachment
attachment = {
"destinationType": "DOWNLOAD",
"destinationDatasetProjectKey": "DKU_CHURN",
"overwriteDestinationDataset": "false",
"selection": {
"samplingMethod": "FULL",
"partitionSelectionMethod": "ALL",
"targetRatio": 0.02,
"maxRecords": 100000,
"selectedPartitions": [],
"ordering": {
"enabled": "false",
"rules": []
}
},
"advancedMode": "false",
"exportOption": {
"id": "excel",
"label": "Excel (*.xlsx)",
"canStream": "false",
"formatType": "excel",
"predefinedConfig": {
from dataiku.scenario import Scenario scenario = Scenario() # Note that you must be admin to update global variables scenario.set_project_variables(var1="val1", var2=3)
from dataiku.scenario import Scenario scenario = Scenario() scenario.set_scenario_variables(var1="val1", var2=34)
from dataiku.scenario import Scenario
scenario = Scenario()
scenario.synchronize_hive_metastore("mydataset")
from dataiku.scenario import Scenario
scenario = Scenario()
scenario.build_dataset("mydataset")