Giters

MIC-DKFZ / nnDetection

nnDetection is a self-configuring framework for 3D (volumetric) medical object detection which can be applied to new data sets without manual intervention. It includes guides for 12 data sets that were used to develop and evaluate the performance of the proposed method.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

[Question] About prediction

yl3639 opened this issue · comments

commented

❓ Question

Hi,
May I ask is there any way to speed up the predictions? I have images with resolutions about(500, 500, 300) and the prediction time of a case is about 4mins.

Thanks!

image
commented

Hi,

there are several parameters which can be adjusted to speed up the inference but smaller performance degredations are to be expected there (here in order from [likely] lowest perf impact to highest):

  • disable test time augmentation
  • use less models (inference only)
  • decrease patch overlap while predicting

Best,
Michael

commented

This issue is stale because it has been open for 30 days with no activity.

commented

This issue was closed because it has been inactive for 14 days since being marked as stale.