Deep Learning Specialization: Structuring Machine Learning Projects: study plan and how long it takes
The DeepLearningAI playlist is 22 videos and 3.4 hours long. Here's how long it takes at different paces, a week-by-week schedule built from the real videos, and a planner that turns it into your own daily plan.
22 videos · 3.4 hours · DeepLearningAI
Course 3 of the Deep Learning Specialization is about strategy rather than algorithms: single-number evaluation metrics, setting up dev and test sets, avoidable bias and human-level performance, error analysis, building a first system quickly, mismatched data distributions, transfer learning, multi-task learning and end-to-end deep learning.
It's short, at under four hours, and widely considered one of the most practical ML courses available.
Who it's for: Anyone building ML systems who wants a framework for deciding what to try next.
How long does it take?
The full playlist is 3.4 hours across 22 videos. Here's roughly how many days that is at different daily watch times and speeds, counting video time only:
| Daily time | At 1x | At 1.5x | At 2x |
|---|---|---|---|
| 30 min a day | 7 days | 5 days | 4 days |
| 1 h a day | 4 days | 3 days | 2 days |
| 2 h a day | 2 days | 2 days | 1 days |
Make your own schedule
The playlist link is already filled in. Press Plan, set how much time you have each day and your speed, and you'll get a day-by-day schedule with a finish date and a calendar file. It reads the live playlist, so new videos are included.
Week-by-week schedule
A sample plan at 30 minutes a day and normal speed, using whole videos so none is split across days. That's 2 weeks in total. For a different pace, start date or speed, use the planner above.
Week 1Videos 1–20 · 3h
Day 1 · 27m
- 1Improving Model Performance (C3W1L01)3m
- 2Orthogonalization (C3W1L02 )11m
- 3Single Number Evaluation Metric (C3W1L03)7m
- 4Satisficing and Optimizing Metrics (C3W1L04)6m
Day 2 · 29m
- 5Train/Dev/Test Set Distributions (C3W1L05)7m
- 6Sizeof Dev and Test Sets (C3W1L06)6m
- 7When to Change Dev/Test Sets (C3W1L07)11m
- 8C3W1L08 WhyHumanLevelPerformance6m
Day 3 · 25m
- 9Avoidable Bias (C3W1L09)7m
- 10Understanding Human-Level Performance? (C3W1L10)11m
- 11Surpassing Human-Level Performance (C3W1L11)6m
Day 4 · 30m
- 12Improving Model Performance (C3W1L12)6m
- 13Carrying Out Error Analysis (C3W2L01)11m
- 14Cleaning Up Incorrectly Labelled Data (C3W2L02)13m
Day 5 · 17m
- 15Build First System Quickly, Then Iterate (C3W2L03)6m
- 16Training and Testing on Different Distributions (C3W2L04)11m
Day 6 · 28m
- 17Bias and Variance With Mismatched Data (C3W2L05)18m
- 18Addressing Data Mismatch (C3W2L06)10m
Day 7 · 24m
- 19Transfer Learning (C3W2L07)11m
- 20Multitask Learning (C3W2L08)13m
Week 2Videos 21–22 · 22m
Day 8 · 22m
- 21What is end-to-end deep learning? (C3W2L09)12m
- 22Whether to Use End-To-End Deep Learning (C3W2L10)10m
Tips for finishing
- Watch it in a week, then apply the error analysis process to a project of your own.
- Pair it with CS229's lecture on debugging ML models; they make the same points from different angles.