README.md

Machine Learning Projects

Repository contains ML-related code:

1. ML util functions:
   1. common/
   2. clustering/
   3. analysis/
   4. preprocessing/
   5. timeseries/
2. Recent ML projects:
   1. otto/: Otto project from Kaggle
   2. cv_course_udacity/: code for projects on Computer Vision Nanodegree on Udacity
   3. airbus_ship_detection/: project from Kaggle For more details see README files in each of the folders for more details.

More of my ML projects are in separate repos:

1. ETH Data mining course projects
2. ETH Machine learning course projects
3. ETH CIL Road segmentation project
4. ETH Computer vision course exercises

Tests

From the this directory run python -m unittest discover

TODO: add tests for KMedoids

Useful packages and liks:

• stacking and boosting of models: vecstack
• model-based hyper-parameter search: scikit-optimize
• Overview of feature encoding techniques