https://github.com/miguellopezvirues/azure_keyword_cpc

Development and deployment of simple regression model in Azure Machine Learning.
https://github.com/miguellopezvirues/azure_keyword_cpc

azureml deplyment machine-learning mlflow pandas scikit-learn

Last synced: 30 days ago
JSON representation

Development and deployment of simple regression model in Azure Machine Learning.

• Host: GitHub
• URL: https://github.com/miguellopezvirues/azure_keyword_cpc
• Owner: MiguelLopezVirues
• Created: 2024-05-31T11:02:13.000Z (8 months ago)
• Default Branch: main
• Last Pushed: 2024-06-02T12:17:22.000Z (8 months ago)
• Last Synced: 2024-11-06T06:13:22.565Z (3 months ago)
• Topics: azureml, deplyment, machine-learning, mlflow, pandas, scikit-learn
• Language: Jupyter Notebook
• Homepage:
• Size: 5.88 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# azure_keyword_CPC

This project consists in the development and deployment of a CPC prediction model using Python as programming language, MLFlow as the framework for experiment monitoring and model logging, and Azure Machine Learning for data asset, compute, training and deployment management. 

## Structure 

- KW_CPC_development.ipynb: Data exploration and development machine learning model. 

- KW_CPC_pipeline_training_deployment.ipynb: Creation of pipeline components to prepare the data and train the model. Model log, registration and online deployment for realtime prediciton.

- prepare-data.yml: Pipeline component for data preparation.

- train-model.yml: Pipeline component for training and model log.

- /src/: Folder with generated scripts called by the yaml pipeline components.

- /data/: Folder with dataset and sample data for deployment invoke.

- /artifacts/: Model and execution logs from the pipeline job.

- /named-outputs/: Outputs uploaded by outputs mode of the pipeline.

-----

Used as a practise for the certification DP-100, obtained on the 20th April 2024.