choderalab / modelforge

Infrastructure to implement and train NNPs
https://modelforge.readthedocs.io/en/latest/
MIT License
9 stars 4 forks source link

Initial dataset structure #3

Closed wiederm closed 11 months ago

wiederm commented 11 months ago

Description

Outline the basic dataset structure

This PR contains the structure for a BaseDataset that inherits from torch.utils.data.Dataset and contains functions to retrieve, manipulate and cache the downloaded dataset. It also contains an implementation of a QM9Dataset that provides functions that control from which remote location the dataset is retrieved and the transformations to generate an opimized npy file. It also implements the __getitem__ and __len__ calls necessary for the torch.utils.data.Dataset class.

Todos

Status