2018 Data Science Bowl

https://www.kaggle.com/c/data-science-bowl-2018 My approach to 2018 Data Science Bowl. The object is to find the nuclei in divergent images to advance medical discovery. We use a U-Net (https://arxiv.org/abs/1505.04597) architecture which is normally using for segmentation tasks in biomedical imaging.

Name		Name	Last commit message	Last commit date
Latest commit History 14 Commits
__pycache__		__pycache__
data		data
logs		logs
saves		saves
.DS_Store		.DS_Store
README.md		README.md
U_Net_Model.py		U_Net_Model.py
checkpoint		checkpoint
load_data.py		load_data.py
main.py		main.py
metrics.py		metrics.py
nn0_256_256_3.data-00000-of-00001		nn0_256_256_3.data-00000-of-00001
nn0_256_256_3.index		nn0_256_256_3.index
nn0_256_256_3.meta		nn0_256_256_3.meta
predictions.py		predictions.py
sub-dsbowl2018-1.csv		sub-dsbowl2018-1.csv
transform_data.py		transform_data.py

mfournarakis/ScienceBowl