README

Project results for Advanced Machine Learining course. Innopolis UNiversity 2019.

1.   amino_dgl_build_graph.ipynb - pipeline for data preprocessing (NOT FOR DEMO!)
1.1. amino_acid_prop.csv - service file for data preprocessing
1.2. graphs_amino.pickle - results of data preprocessing - list of graphs saved in DGL structure
1.3  lables_amino.pickle - results of data preprocessing - list of labels

2. GIN - The first model based on Xu K., How Powerful are Graph Neural Networks? (https://arxiv.org/abs/1810.00826)
2.1 GIN_learn.ipynb - demo file with GIN model learning 

3. Hierarchical - The second model based on Rex Y., Hierarchical Graph Representation Learning with Differentiable Pooling (https://arxiv.org/abs/1806.08804)
3.1 Hierarchical_learn.ipynb - demo file with Hierarchical model learning