Contents

Preliminary ProgramΒΆ

This section lists the techniques that are exposed to the participants. It is worth noting that a variety of techniques are introduced in order to give the participants a general look on the existing technologies but not all of these techniques are used in the practical use cases.

  1. Data preparation

    • Data cleaning and missing data reconstruction

    • Mean imputation

    • Interpolation

    • K-nearest neighbors reconstruction

    • Image inpainting

  2. Data visualization

    • Matplotlib

    • Plotly

    • Dash

  3. Feature extraction using dimensionality reduction techniques (unsupervised)

    • Principal Component Analysis (PCA) ==> unsupervised

    • Variational Autoencoder ==> unsupervised

  4. Data clustering (unsupervised)

    • K-means

    • Mini-batch K-means

    • Super-pixel segmentation

  5. Classification, Segmentation and Regression (supervised)

    • Random forest

    • XGBoost

    • LSTM

    • Convolutional networks

  6. Model Evaluation

    • Precision

    • Recall

    • F1-score

    • Confusion matrix

  7. Hyperparameter tuning

    • Learning curves

    • Random search

    • Grid search