def load(self):
content = {}
if hasattr(self.ctx, 'credentials'):
content = self.ctx.credentials
elif 'AZURE_CREDENTIALS' in os.environ:
content = os.environ.get('AZURE_CREDENTIALS', None)
content = json.loads(content) if content else {}
else:
#look for service principal credentials
if self._credentials_file_exist():
with open(self.creds_file, 'r') as file:
content = json.loads(file.read())
else:
content = self._load_azure_cred_file()
if not content.get('subscription_id'
) and not self.ctx.config.get('use_server'):
from azureml.core.authentication import InteractiveLoginAuthentication
#fallback to force browser login for token
interactive_auth = InteractiveLoginAuthentication(
force=True)
interactive_auth.get_authentication_header()
content = self._load_azure_cred_file()
self.subscription_id = content.get('subscription_id')
self.directory_tenant_id = content.get('directory_tenant_id')
self.application_client_id = content.get('application_client_id')
self.client_secret = content.get('client_secret')
return self
def query_confidence(query):
# Get a token to authenticate to the compute instance from remote
interactive_auth = InteractiveLoginAuthentication()
auth_header = interactive_auth.get_authentication_header()
headers = auth_header
# Add content type header
headers.update({'Content-Type': 'application/json'})
# Sample data to send to the service
test_sample = bytes(query, encoding='utf8')
# Replace with the URL for your compute instance, as determined from the previous section
service_url = "https://fact-check-8890.eastus.instances.azureml.net/score"
# for a compute instance, the url would be https://vm-name-6789.northcentralus.instances.azureml.net/score
resp = requests.post(service_url, test_sample, headers=headers)
#print("prediction:", resp.text)
response = [float(i) for i in list(((resp.text[2:])[:-2]).split(','))]
return response
# COMMAND ---------- # MAGIC %run ./99-Shared-Functions-and-Settings # COMMAND ---------- # MAGIC %md # MAGIC #### Login to Azure Machine Learning Workspace # COMMAND ---------- from azureml.core import Workspace, Experiment, Run from azureml.core.authentication import InteractiveLoginAuthentication up = InteractiveLoginAuthentication() up.get_authentication_header() ws = Workspace(**AZURE_ML_CONF, auth=up) # COMMAND ---------- from pyspark.ml import Pipeline from pyspark.sql.functions import * from pyspark.ml.feature import StringIndexer, OneHotEncoderEstimator, VectorAssembler, StandardScaler from pyspark.ml.regression import RandomForestRegressor, LinearRegression, GBTRegressor from pyspark.ml.evaluation import RegressionEvaluator import matplotlib.pyplot as plt import shutil import os # COMMAND ----------
published_pipeline = succ_pipeline.publish(
name="succ_Planning_Reg_Sevice_Model3",
description="Trains model and Creates the Service",
version="1.0")
rest_endpoint = published_pipeline.endpoint
print(rest_endpoint)
from azureml.core.authentication import InteractiveLoginAuthentication
interactive_auth = InteractiveLoginAuthentication()
auth_header = interactive_auth.get_authentication_header()
import requests
experiment_name = 'RegExp3'
response = requests.post(rest_endpoint,
headers=auth_header,
json={
"ExperimentName": experiment_name,
"ParameterAssignments": {
"data": default
}
})
run_id = response.json()["Id"]
run_id