datacamp studies

chapter 3: decorrelating your data and dimension reduction

• visualising the pca transformation
  • dimension reduction removes less-informative features and finds patterns in data to use patterns to re-express it in a compressed form
  • pca (principal component analysis) - two steps of dimesnion reduction technique
    • decorrelation
    • reduces dimension
  • intrinsic dimension
    • the number of features needed to approximate a dataset
    • let’s us know how much a dataset can be compressed
    • can be identified by pca
  • dimension reduction with pca
    • pca features are in decreasing order of variance
    • must specify how many features to keep

chapter 4: discovering interpretable features

• non-negative matrix factorisation (nmf)
  • a dimension reduction technique
  • easy to interpret
  • can only be implemented on a dataset that have sample features that are non-negative

language trilogy

started my maru hiragana drills at 14.08 and finished them in 3 minutes

14.22 - finished half of chapter 4 busuu german

14.48 - finished half of chapter 4 busuu french

15.55 - finished chapter 9 busuu japanese