This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(76.8, 94%, 17%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EGT%3C/text%3E%3C/svg%3E)
I'm currently trying to create an endpoint in Azure ML studio designer, I'm trying to download a ML model pickle file from a blob container and use it in the pipeline to make predictions in new data. But when I try to download the file and load into the script I always get an error. Here follows the code snippet I'm trying to run and the error returned.
Code Snippet
origin = dataframe1['Key'].unique()[0].lower()
run = Run.get_context(allow_offline = True)
ws = run.experiment.workspace
datastore = Datastore(ws, 'models_datastore')
datastore.download('Downloads//', prefix = 'Model_{origin}//vectorizer.pkl')
model = pickle.load(open('Downloads//vectorizer.pkl', 'rb'))
Error returned
Got exception when invoking script at line 23 in function azureml_main: 'FileNotFoundError: [Errno 2] No such file or directory: 'Downloads//vectorizer.pkl''.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(22.400000000000002, 11%, 12%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ERM%3C/text%3E%3C/svg%3E)
@Guilherme Takata Thanks, Can you please confirm are you trying customized Deployment pipeline?. Can you please add more details about the deployment that you are trying.
you could use the ModuleStep class to reference a Designer Module by name: https://github.com/Azure/MachineLearningNotebooks/blob/master/how-to-use-azureml/machine-learning-pipelines/intro-to-pipelines/aml-pipelines-how-to-use-modulestep.ipynb
@Guilherme Takata Thanks for the question. Here are the instructions to Use the studio to deploy models trained in the designer - Azure Machine Learning | Microsoft Learn and document that explains how we can get access to score.py and conda_env.yaml files under Output + logs tab for Train module.