Side Quest
11
Hyperparameter tuning
NLP: From Simple to Spectacular
Welcome
1
Machine learning needs data
Classification
2
Baseline: gotta start somewhere
3
OneR
Representation learning
4
OneR, TwoR, RedR, BlueR: Inputs matter
Classification
5
More rules with Decision Trees
6
What’s in a loss?
Representation learning
7
Words > characters
8
[‘T’, ‘o’, ‘k’, ‘e’, ‘n’, ‘s’]
Generative AI
9
Generating gibberish with unigrams
10
Why one when you can have N?
Side Quest
11
Hyperparameter tuning
Bonus chapters
12
Cleaning data
13
Quality vs. quantity
References
Side Quest
11
Hyperparameter tuning
11
Hyperparameter tuning
TODO
10
Why one when you can have N?
12
Cleaning data