ML Flashcard completion and study

Starsa / Mentorship

Projects for Mentorship

0 stars 0 forks source link

ML Flashcard completion and study #27

Closed Starsa closed 3 years ago

Starsa commented 3 years ago

Discuss flashcards reviewed and list any questions.

[x] Monday
[x] Tuesday
[x] Wednesday
[x] Thursday
[x] Friday
[x] Saturday
[x] Sunday

Starsa commented 3 years ago

ADA Boost vs. Gradient boost.

AdaBoost is the first designed boosting algorithm with a particular loss function. Gradient Boosting is a generic algorithm that assists in searching the approximate solutions to the model.

Gradient Boost is more flexible than AdaBoost.

Starsa commented 3 years ago

Gradient

The gradient of a function is a vector containing all its partial derivatives at some point (repeat I know, just trying to focus on the wording)

Starsa commented 3 years ago

Adjusted R2

A modified version of R-squared that has been adjusted for the number of predictors in the model.
It increases when the new term improves the model more than by chance.
It decreases when a predictor improves the model by less than expected.
Once all correct features have been added, additional features should be peanalized

Starsa commented 3 years ago

Agglomerative Clustering

All observations start as their own cluster
Clusters that meet a criteria are merged
Process is repeated, growing clusters, until some endpoint is met

Starsa commented 3 years ago

AIC

Used to compare models
- like in feature selection

CLuiz commented 3 years ago

This is great! Good fodder for interview questions. AUC is one I typically poke at. R squared and adjusted R squared as well