JSON Wrangling
musicbrains doesn't have API key, so we free to search the music wiki through their API/metadata
CSV Wrangling
- In liine 15, error will show up because it show 2 field with comma delimiter
Fundamentals
- every Data that handled by humans have possibility of errors in it
Bayesian Learning
Today i want to talk about bayesian learning. Bayes' theorem has been around for a while. The name itself is after the statiscian Thomas Bayes. Interesting enough, He died before even published his major scientific achievement to the world as we know today.
Problem Set 2
Bayesian Inference
- Marginalization: to put X, simply count what's not y, and what's y and x, that will produce x
Joint Distribution
- Probabilistic quantities, represent and reason? Called Bayesian Networks
Infinite Hypothesis Spaces
How to count m if the set of hypothesis is infinity?
- VC stated that eventhough the hypothesis space is infinite, as long as the sample is finite, then we can know how much the samples do we need.
Pac Learning
- training error: the training example missclasfied by h (true hypothesis if got none)
Learning Theory
takes closest neighbors
- The first one is decision trees. First it split the region 1/3 left, 2/3 right. then make a horizontal split. and finally another vertical split. So it's splitting into two region, on and on.