FAndrades_Science4Cast

Python implementation of Science4Cast Submission

Autor: Francisco Andrades - Computer Science Engineering student, Federico Santa María technical University - Chile

Email: francisco.andrades@sansano.usm.cl

local_subgraphs.ipynb: Extract local subgraphs.

• Input: Data provided for the Science4Cast challenge: https://drive.google.com/file/d/1Hm7WgT763bIteXcvQDEo_XI0x9gov87y/view?usp=sharing
• Output: Local subgraphs Dataset

gnn-sci4cast.ipynb: Gnn training

• Input: local_subgraphs.ipynb output
  • Dataset for test set: https://drive.google.com/file/d/1W-kGrW8sYGpFJStWVh7-VvMSD18EWAXO/view?usp=sharing
  • Dataset for challenge set: https://drive.google.com/file/d/1NFrez0n_tmD8B1wkPZWpcVZfrf250r5a/view?usp=sharing
• Output: predictions

Please change the variable path in both notebooks according to your directory structure.