search

uci-cbcl / NoduleNet

[MICCAI' 19] NoduleNet: Decoupled False Positive Reduction for Pulmonary Nodule Detection and Segmentation
Other
187 stars 58 forks source link

missing codes #25

Closed LXYTSOS closed 4 years ago

LXYTSOS commented 4 years ago

where is display function in visualize.py and average_precision in test.py, and how to visualize the result as gif file?

tanghaotommy commented 4 years ago

I have uploaded some utility functions in the util.py. Please see the latest commit. For visualization purpose, we first use https://github.com/uci-cbcl/NoduleNet/blob/master/utils/visualize.py#L302 to generate png for each slice. And then use linux command from here https://unix.stackexchange.com/questions/24014/creating-a-gif-animation-from-png-files to convert pngs to gif.

joaco18 commented 4 years ago

Hi I've noticed that in the function 'utils/util.py/average_precision' the precision is calculated this way: p = tp / float(tp + fp + fn). Is this right? Shouldn't it be p = tp / float(tp + fp)' instead?