https://github.com/lukoshkin/ml2020

https://github.com/lukoshkin/ml2020

causal-networks dilated-cnn persistence-landscape sequential-autoencoder tda time-series-classification

Last synced: 7 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/lukoshkin/ml2020
• Owner: lukoshkin
• Created: 2020-04-04T16:02:50.000Z (almost 5 years ago)
• Default Branch: master
• Last Pushed: 2022-12-08T03:59:02.000Z (about 2 years ago)
• Last Synced: 2024-11-07T03:19:23.943Z (about 2 months ago)
• Topics: causal-networks, dilated-cnn, persistence-landscape, sequential-autoencoder, tda, time-series-classification
• Language: Jupyter Notebook
• Size: 83 MB
• Stars: 0
• Watchers: 2
• Forks: 1
• Open Issues: 11
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Skoltech ML course, 2020

***Some implemented features that may become useful later:***

**Homeworks**

* HW2, part1: 

    - plotting with customizable layout (HW2, part1), 

    `plt.GridSpec` from `matplotlib.pyplot`

    - highlighting cells of pd.DataFrame object, 

    `.style.apply(color_fn)`

    - example of two aligned horizontal barplots

    with adjusted tick labels

* HW2, part2:

    - theory problems

* HW3, part1:

    - nothing

* HW3, part2:

    - a bit of theory, just a smattering

*There were 4 homeworks; 

HW1 and HW4 contained nothing useful, so I didn't upload them*

**Project**

*See the project description in

[the project folder](https://github.com/lukoshkin/ml2020/tree/master/project)*

* Exponentially dilated causal convolutions

* Recurrent and convolutional encoder for time series

* Recurrent and convolutional decoder for time series

* Convolutional classifier for time series

* Convolutional Regressor for time series reconstruction

  from persistence landscapes

* Oversampling technique (implemented in PyTorch)

* Some other tools