Building a Reproducible Model Workflow

Brought by: Udacity

Overview

Learn to be more productive through ML projects that require reproducible workflow best practices.

Syllabus

  • Machine Learning Pipeline
    • Learn MLOps fundamentals and dive into version data and artifacts. Write a ML pipeline component and link together ML components.
  • Data Exploration & Preparation
    • Execute and track the Exploratory Data Analysis (EDA). Clean and pre-process the data and segregate (split) datasets.
  • Data Validation
    • Use pytest with parameters for reproducible and automatic data tests. Perform deterministic and non-deterministic data tests.
  • Training, Validation & Experiment Tracking
    • Tame the chaos with experiment, code, and data tracking. Track experiments with W&B. Validate and choose the best-performing model. Export model as an inference artifact and test final inference artifact.
  • Release & Deploy
    • Release pipeline code and learn options for deployment and how to deploy a model.
  • Course Project: Build an ML Pipeline for Short-Term Rental Prices in NYC
    • Write a machine learning pipeline to solve the following problem: A property management company is renting rooms and properties in New York for short periods on various rental platforms. They need to estimate the typical price for a given property based on the price of similar properties. The company receives new data in bulk every week, so the model needs to be retrained with the same cadence, necessitating a reusable pipeline. Write an end-to-end pipeline covering data fetching, validation, segregation, train and validation, test, and release. Run it on an initial data sample, then re-run it on a new data sample simulating a new data delivery.

Taught by

nd0821 Giacomo Vianello

Building a Reproducible Model Workflow
Go to course

Building a Reproducible Model Workflow

Brought by: Udacity

  • Udacity
  • Free
  • English
  • Certificate Not Available
  • Certain days
  • advanced
  • N/A