Link Examples

Jupyter notebooks in this folder illustrate how to construct Link^TM pipelines when conducting exploratory data analysis or performing machine learning modelling task.

1. Spiral Pattern Classification (notebook, screenshot)
2. Iris Data EDA and Modelling (notebook, screenshot)
3. Titanic Data EDA and Modelling (notebook, screenshot)
4. Image Generation using Variational Autoencoder (notebook, screenshot)
5. Image Restoration using Denoising Autoencoder (notebook, screenshot)
6. MNIST Image Classification using CNN (notebook, screenshot)
7. Using XGBoost for scikit-learn Datasets (notebook, screenshot)
8. Text Data Classification using RNN (notebook, screenshot)
9. Deep Q-Network Reinforcement Learning for CartPole Environment (notebook, screenshot)

How to view a Link example notebook

1. Clone this repository to get all the example notebooks (git clone https://github.com/makinarocks/link-example).
   Or download a notebook file you want to view as follows:
   • Click the notebook's link.
   • In the github-rendered page, click the mouse right button on the raw button and select "Save As..." menu.
   • Save the file as Jupyter notebook.
2. If you dont't have Link^TM installed on your local machine, first install the program. Visit Link^TM homepage for more information. (https://link.makinarocks.ai/)
3. Run Link^TM by executing jupyter lab on your terminal.
4. Open the downloaded notebook file.

The procedure above is roughly illustrated in the motion GIF below.

How to use Link

For more information on user guide of Link^TM, please visit the introduction page at https://makinarocks.gitbook.io/link/v/en/.

---

Spiral Pattern Classification

Classification of 2D spiral-distributed data using Pytorch framework

Iris Data EDA and Modelling

Basic exploratory data analysis (EDA) and modelling of iris data using Scikit-Learn library

Titanic Data EDA and Modelling

Various EDA and ensemble modelling of titanic data using Scikit-Learn library

Image Generation using Variational Autoencoder

MNIST image generation test with variational autoencoder (VAE) using Pytorch framework

Image Restoration using Denoising Autoencoder

Restoring corrupted MNIST images with denoising autoencoder using Pytorch framework

MNIST Image Classification using CNN

Classification of MNIST images with convolutional neural network (CNN) and fully-connected network (FCN) being compared using Pytorch framework

Using XGBoost for scikit-learn datasets

Modelling of diabetes data for regression, classificaion, cross-validation, and hyperparameter searching using Scikit-Learn and XGBoost library

Text Data Classification using RNN

Text classification with RNN using Pytorch library

Deep Q-Network Reinforcement Learning for CartPole Environment

Deep Q-Network Reignforcement Learning (RL) for CartPole environment using Pytorch and OpenAI-Gym frameworks