Discover some of the reasons why some algorithms are much slower than others while focusing on fitting a large number of coefficients.
- [Instructor] The second reason…that some algorithms are much slower than others…is some algorithms just perform more calculations.…Let's take a look.…This is an artificial neural network…that's been built on the same data set.…All we're trying to do is predict whether…or not an expectant mom will have a low birth weight baby.…Now, artificial neural networks get quite technical,…but take a look.…I just want you to attend to the very large number…of lines.…If we were using a statistically-based approach,…we've only got three variables here,…weight of the mom, uterine irritability, and hypertension.…
A statistical approach would only do four calculations,…one per variable and what's called the constant,…but the point is there would only be four.…But look at all these lines, many, many more,…so the neural net is doing many more calculations.…That makes it more accurate, but it makes it slower.…Over here, we've got six times five lines,…and yes, indeed, there's a relationship between the number…of shapes on the left and the number…
Author
Released
12/11/2018Note: This course is software agnostic. The emphasis is on strategy and planning. Examples, calculations, and software results shown are for training purposes only.
- Evaluating the proper amount of data
- Assessing data quality and quantity
- Seasonality and time alignment
- Data preparation challenges
- Data modeling challenges
- Scoring machine-learning models
- Deploying models and adjusting data prep and scoring
- Monitoring and maintenance
Skill Level Beginner
Duration
Views
Related Courses
-
Machine Learning and AI Foundations: Recommendations
with Adam Geitgey58m 7s Intermediate -
Deploying Scalable Machine Learning for Data Science
with Dan Sullivan1h 43m Intermediate
-
Introduction
-
Defining terms1m 48s
-
1. The Phases of a Machine Learning Project
-
2. Designing a Machine Learning Dataset
-
How much data do I need?2m 12s
-
Balancing1m 56s
-
Who truly has big data?3m 43s
-
Assessing data3m 32s
-
-
3. Data Prep Challenges
-
Data and the data scientist2m 55s
-
Dummy coding2m 1s
-
Feature engineering2m 51s
-
-
4. Modeling Challenges
-
Slow algorithms: Brute force1m 59s
-
Slow algorithms: More models2m 24s
-
How to sample properly2m 36s
-
Modeling with missing data3m 37s
-
5. Scoring
-
Scoring a black box model2m 50s
-
Scoring an ensemble1m 49s
-
6. Deployment
-
Batch vs. real-time scoring4m 39s
-
Data prep and scoring2m 59s
-
-
7. Monitoring and Maintenance
-
Conclusion
-
Next steps1m 1s
-
- Mark as unwatched
- Mark all as unwatched
Are you sure you want to mark all the videos in this course as unwatched?
This will not affect your course history, your reports, or your certificates of completion for this course.
CancelTake notes with your new membership!
Type in the entry box, then click Enter to save your note.
1:30Press on any video thumbnail to jump immediately to the timecode shown.
Notes are saved with you account but can also be exported as plain text, MS Word, PDF, Google Doc, or Evernote.
Share this video
Embed this video
Video: Slow algorithms: More calculations