MIC-DKFZ / medicaldetectiontoolkit

The Medical Detection Toolkit contains 2D + 3D implementations of prevalent object detectors such as Mask R-CNN, Retina Net, Retina U-Net, as well as a training and inference framework focused on dealing with medical images.
Apache License 2.0
1.31k stars 297 forks source link

Interpolate in backbone.py #69

Closed yoshua133 closed 5 years ago

yoshua133 commented 5 years ago

Hi, I want to point out an error: In the backbone.py following lines

if conv.dim == 2: self.P1_upsample = Interpolate(scale_factor=2, mode='bilinear') self.P2_upsample = Interpolate(scale_factor=2, mode='bilinear') else: self.P1_upsample = Interpolate(scale_factor=(2, 2, 1), mode='trilinear') self.P2_upsample = Interpolate(scale_factor=(2, 2, 1), mode='trilinear')

You didn't mention where is the Interpolate from, so maybe you could write it more specifically.

Thank you

pfjaeger commented 5 years ago

See my comment on your conv class post...