search

wentaozhu / DeepLung

WACV18 paper "DeepLung: Deep 3D Dual Path Nets for Automated Pulmonary Nodule Detection and Classification"
Apache License 2.0
316 stars 143 forks source link

run frocwrtdetpepchluna16.py with tianchi dataset #36

Closed leowangzi closed 6 years ago

leowangzi commented 6 years ago

The Tianchi dataset did not generate _extendbox.npy in prepare.py. in frocwrtdetpepchluna16.py, the convertcsv function need to load _extendbox.npy and do some matrix operations like this:

Pbb[:, 1:] = np.array(pbb[:, 1:] + np.expand_dims(extendbox[:,0], 1).T)
Pbb[:, 1:] = np.array(pbb[:, 1:] * np.expand_dims(resolution, 1).T / np.expand_dims(spacing, 1).T)

How to deal with this step? thanks.

wentaozhu commented 6 years ago

frocwrtdetpepchluna16.py is to calculate the FROC. You need first train the model, then test. Then calculate the FROC. prepare.py is the pre-processing.

leowangzi commented 6 years ago

@wentaozhu I have finished the training and the test. I met the above problems when calculating FROC.

leowangzi commented 6 years ago

newplot

leowangzi commented 6 years ago

default

wentaozhu commented 6 years ago

Your loss curve seems weird. Maybe you can first try Luna 16.