Skip to content
/ SDATA Public

A compilation of small projects during SDATA courses

Notifications You must be signed in to change notification settings

BDouchet/SDATA

Repository files navigation

SDATA

TP_LDA

Work on iris dataset for classification using Linear Discriminant Analysis. In this project :

  • Data Visualization (pairplot, TSNE)
  • Calculation of the LDA manually
  • Comparison with sklearn LDA
  • Training phases with train_test_split or K-fold and comparison with f1_score
  • Display of decision boundaries

TP_Regularization

Work on house price dataset to estimate the price of an house according to its description.

  • Data visualization, Preprocessing and Data Cleaning
  • Single Linear Regression
  • Multiple Linear Regression
  • Ridge Regression (L2 Regularization)
  • Lasso Regression (L1 Regularization)

NeuralNetworks

Work on classical CNNs on MNIST and CIFAR10

  • Data Preprocessing and Visualization
  • Build and Training of CNNs
  • Feature Extraction
  • Comparison to trees

Kalman

Work on Linear Gaussian Model (state space model) to apply Kalman Filter :

  • Mathematical Background
  • Simulation of the state space model
  • Application of a Kalman Filter
  • Exploration with different initial parameters

About

A compilation of small projects during SDATA courses

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published