Learn Before
An Iterative Machine Learning Workflow
Machine learning development usually proceeds in cycles. A practical loop is to propose an approach, implement it, run an experiment to measure the result, and then use what you learned to choose the next change. Because each round reveals new information, teams often test many ideas before settling on one that works well.
0
1
Tags
D2L
Dive into Deep Learning @ D2L
Machine Learning
Deep Learning
Supervised Learning
Data Science
Machine Learning Strategy
Machine Learning Yearning @ DeepLearning.AI
Related
Representational Learning
Supervised Learning
Kaggle Platform
Predictive Analytics for Accelerated Decision-Making
Development Set
Purpose of a Test Set
Combining Metrics With a Threshold and an Objective
It Is Hard to Predict the Best Machine Learning Strategy Before Trying It
An Iterative Machine Learning Workflow
Bias and Variance as Two Major Sources of Error
What End-to-End Learning Can Predict
When a Pipeline Is Missing Information
Text Polarity Detection
Machine Learning Strategy
Two Forces Behind Recent Deep Learning Gains
Learn After
Faster Experiment Cycles Improve Learning Speed
Evaluate Several Model Improvements at the Same Time
Share Machine Learning Lessons With the Team
What three-step cycle best describes a fast machine-learning development loop?
Does the first design you test in an ML project usually become the final solution?
Model Improvement Usually Takes Many Iterations
Match each stage in a simple development cycle with its role.
Put the machine learning improvement cycle in the right order.
Why Machine Learning Development Repeats in Cycles
What should a team do after the first experiment makes things worse?
What should happen after an experiment ends?
What is the main purpose of running a trial in an iterative machine learning workflow?
Must you often test many ideas before finding a workable machine learning solution?