Skip to content

Files

Latest commit

Oct 2, 2018
8995e7b · Oct 2, 2018

History

History
14 lines (11 loc) · 688 Bytes

README.md

File metadata and controls

14 lines (11 loc) · 688 Bytes

MARCO Dataset

Classification of crystallization outcomes using deep convolutional neural networks. The Machine Recognition of Crystallization Outcomes (MARCO) initiative has assembled roughly half a million annotated images of macromolecular crystallization experiments from various sources and setups. The datasets contains images belonging to either of 4 classes i.e., Clear, Precipitate, Crystals or Other.

Classes

0 - Clear
1 - Crystals
2 - Other
3 - Precipitate

Links

Arxiv : https://arxiv.org/pdf/1803.10342.pdf
Dataset : https://marco.ccr.buffalo.edu/download