jwyang/fpn.pytorch

Issues
Search Issues
Labels Milestones New issue
Search results
Open
Closed
ImportError:/fpn.pytorch-master/lib/model/nms/_ext/nms/_nms.so: undefined symbol: __cudaPopCallConfiguration
Status: Open.
#56 In jwyang/fpn.pytorch;
· AndrewNest opened on May 8, 2021
sh make.sh 编译不通过
Status: Open.
#55 In jwyang/fpn.pytorch;
· lllllmaster-lulllll opened on Nov 29, 2020
Does it support PyTorch1.0 or higher?
Status: Open.
#54 In jwyang/fpn.pytorch;
· yudawen opened on Jun 21, 2020
Why is the result not as good as faster-rcnn? Have you modified the code?
Status: Open.
#53 In jwyang/fpn.pytorch;
· puyiwen opened on Jun 3, 2020
train with multi gpus can not work
Status: Open.
#52 In jwyang/fpn.pytorch;
· mynameischaos opened on May 18, 2020
Conversion to TensorRT
Status: Open.
#51 In jwyang/fpn.pytorch;
· ma-siddiqui opened on Dec 15, 2019
roi pooling
Status: Open.
#50 In jwyang/fpn.pytorch;
· Youggie opened on Nov 30, 2019
torch.log(torch.sqrt(h * w) / 224.0) in fpn.py
Status: Open.
#49 In jwyang/fpn.pytorch;
· fafancier opened on Nov 13, 2019
train on my dataset
Status: Open.
#48 In jwyang/fpn.pytorch;
· ttianwang opened on Sep 24, 2019
from model.utils.cython_bbox import bbox_overlaps
Status: Open.
#47 In jwyang/fpn.pytorch;
· brooklyn1900 opened on Sep 13, 2019
about rpn_loss_box is zero
Status: Open.
#46 In jwyang/fpn.pytorch;
· ww451575464 opened on Sep 7, 2019
There was a bug when I continued to train the model. RuntimeError: Expected object of type torch.FloatTensor but found type torch.cuda.FloatTensor for argument #4 'other'
Status: Open.
#45 In jwyang/fpn.pytorch;
· smileSJin opened on Sep 2, 2019

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

ImportError:/fpn.pytorch-master/lib/model/nms/_ext/nms/_nms.so: undefined symbol: __cudaPopCallConfiguration

sh make.sh 编译不通过

Does it support PyTorch1.0 or higher?

Why is the result not as good as faster-rcnn? Have you modified the code?

train with multi gpus can not work

Conversion to TensorRT

roi pooling

torch.log(torch.sqrt(h * w) / 224.0) in fpn.py

train on my dataset

from model.utils.cython_bbox import bbox_overlaps

about rpn_loss_box is zero

There was a bug when I continued to train the model. RuntimeError: Expected object of type torch.FloatTensor but found type torch.cuda.FloatTensor for argument #4 'other'