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
out of memory ConvertSegToBoundingBoxCoordinates #117
I am trying to run MASKRCNN inference on my data, but the process crashes in this line in utils.py:
rois = np.array([(clusters == ii) * 1 for ii in range(1, n_cands + 1)]) # separate clusters and concat
More specifically the process is killed, because of too high memory consumption.
It's because I have a large number of instances per one test volume (around 400) and volumes of sizes 256^3 so the previous line tries to allocate (256^3)*400 booleans.
Do you have any suggestions on how to solve this problem?
Hey,
I am trying to run MASKRCNN inference on my data, but the process crashes in this line in utils.py:
rois = np.array([(clusters == ii) * 1 for ii in range(1, n_cands + 1)]) # separate clusters and concatMore specifically the process is killed, because of too high memory consumption.It's because I have a large number of instances per one test volume (around 400) and volumes of sizes 256^3 so the previous line tries to allocate (256^3)*400 booleans. Do you have any suggestions on how to solve this problem?