cannot import name 'RollingOriginValidator' - azure-machine-learning-service

Question

I'm writing a scoring step with Azure Machine Learning pipeline.
This is the score code:
import os
import pickle
import pandas as pd
from azureml.core.model import Model
import argparse
from azureml.core.run import Run, _OfflineRun
from azureml.automl.core.shared.rolling_origin_validator import RollingOriginValidator
# Called when the deployed service starts
def init():
global model
# Get the path where the deployed model can be found.
run = Run.get_context()
model_path = Model.get_model_path('best_model_data')
print(model_path)
with open(model_path,"rb") as f:
model = pickle.load(f)
print("loaded")
# Handle requests to the service
def run(data):
try:
# Pick out the text property of the JSON request.
# This expects a request in the form of {"text": "some text to score for sentiment"}
prediction = predict(data)
#Return prediction
return prediction
except Exception as e:
error = str(e)
return error
# Predict sentiment using the model
def predict(data, include_neutral=True):
# Tokenize text
test_data_features=data.drop('ProposalId', 1).drop('CombinedTactics',1)
test_data_combos=data['CombinedTactics']
print("data")
# Predict
score = model.predict_proba(test_data_features)
print("predicted")
df=pd.DataFrame({'score':score[:, 1],'CombinedTactics':test_data_combos})
return df
This is the pipeline step definition:
parallel_run_config = ParallelRunConfig(
environment=myenv,
entry_script="use_model.py",
source_directory="./",
output_action="append_row",
mini_batch_size="20",
error_threshold=1,
compute_target=compute_target,
process_count_per_node=2,
node_count=2
)
parallel_step_name = "batchscoring-" + datetime.now().strftime("%Y%m%d%H%M")
batch_score_step = ParallelRunStep(
name=parallel_step_name,
inputs=[test_data.as_named_input("test_data")],
output=output_dir,
parallel_run_config=parallel_run_config,
allow_reuse=True
)
However, I met below error:
File "/mnt/batch/tasks/shared/LS_root/jobs/ucmopp-ws/azureml/70828787-7515-4db4-b448-a5a4b6c0c8ff/mounts/workspaceblobstore/azureml/70828787-7515-4db4-b448-a5a4b6c0c8ff/driver/azureml_user/parallel_run/score_module.py", line 139, in call_init
self.init()
File "/mnt/batch/tasks/shared/LS_root/jobs/ucmopp-ws/azureml/70828787-7515-4db4-b448-a5a4b6c0c8ff/mounts/workspaceblobstore/azureml/70828787-7515-4db4-b448-a5a4b6c0c8ff/use_model.py", line 17, in init
model = pickle.load(f)
File "/azureml-envs/azureml_7e62c7905267a978aa40f8554487e9b9/lib/python3.6/site-packages/azureml/automl/runtime/featurization/init.py", line 8, in
from .data_transformer import DataTransformer, TransformerAndMapper
File "/azureml-envs/azureml_7e62c7905267a978aa40f8554487e9b9/lib/python3.6/site-packages/azureml/automl/runtime/featurization/data_transformer.py", line 54, in
from ..featurizer.transformer import (AutoMLTransformer, CategoricalFeaturizers, DateTimeFeaturesTransformer,
File "/azureml-envs/azureml_7e62c7905267a978aa40f8554487e9b9/lib/python3.6/site-packages/azureml/automl/runtime/featurizer/transformer/init.py", line 28, in
from .timeseries import TimeSeriesTransformer, TimeSeriesPipelineType, NumericalizeTransformer,
File "/azureml-envs/azureml_7e62c7905267a978aa40f8554487e9b9/lib/python3.6/site-packages/azureml/automl/runtime/featurizer/transformer/timeseries/init.py", line 65, in
from azureml.automl.core.shared.rolling_origin_validator import RollingOriginValidator
ImportError: cannot import name 'RollingOriginValidator'
Does anyone have any idea about this error?