During testing DCR-v1, I have problem in getting the document '_detections_all.pkl'. I get it by run
python experiments/faster_rcnn/rcnn_test.py --cfg experiments/faster_rcnn/cfgs/resnet_v1_101_coco_train2017_rcnn_end2end.yaml --thresh 0 --disable-nms
and make some modifications according to my understanding.However,as I run ‘rcnn_rescore_combined_fast.py’, there is memory explosion.
Then in running ‘rcnn_rescore_combined_fast.py’, I try to reduce the parameter(max_per_image) from 100 to 50.And I reduce the image size from 224 to 112.Then I got very terrible results. Could please help me to deal with the problem?
I have one GeForce GTX 1080Ti with 12 GB memory.
During testing DCR-v1, I have problem in getting the document '_detections_all.pkl'. I get it by run
python experiments/faster_rcnn/rcnn_test.py --cfg experiments/faster_rcnn/cfgs/resnet_v1_101_coco_train2017_rcnn_end2end.yaml --thresh 0 --disable-nmsand make some modifications according to my understanding.However,as I run ‘rcnn_rescore_combined_fast.py’, there is memory explosion. Then in running ‘rcnn_rescore_combined_fast.py’, I try to reduce the parameter(max_per_image) from 100 to 50.And I reduce the image size from 224 to 112.Then I got very terrible results. Could please help me to deal with the problem? I have one GeForce GTX 1080Ti with 12 GB memory.