Detectron.pytorch

Detectron.pytorch copied to clipboard

I suppose you just change the repo name at some point. Fix typos.

mimoralea

Convert all Proposal code to PyTorch

1

I have done some profiling and the proposal part of the model (`Generalized_RCNN.RPN`) takes nearly 20% of the time during GPU inference for me. I looked into it and found...

Rizhiy

PR: Add setuptools setup file, add command to compile CUDA/C libraries and general package file refactoring

1

Thanks for your great work and effort in porting Detectron to PyTorch! This PR adds a compatible setuptools distributable setup file, this addition allows other reseachers/developers to embed Detectron in...

andfoy

Cannot run inference in Jupyter

2

@roytseng-tw Hi, I run the inference based on your `infer_simple.py` successfully . At same environment (one gpu, same machine, same folder path), i use it in Jupyter for inference, but...

QiaoranC

Handle NaN loss

2

To avoid division by zero or passing zeros or negative values to log in ``bbox_transform_inv`` I replace those values with ``cfg.EPS``. In addition, the NaN mask loss which rises after...

MFattouh

Is it possible to train Faster-RCNN with resnet50 as backbone with this code?

xuw080

Train from scratch, I got this problem!

1

RuntimeWarning: invalid value encountered in greater_equal (x_ctr < im_info[1]) & (y_ctr < im_info[0]))[0]

RobotIntelligence

I meet problem during implement light_head_rcnn

14

loss_bbox is not converge.other loss(loss_cls,loss_rpn_cls,loss_bbox) is converge.can I push the code to you for debug.

hewumars

Hi, Is there the program or function that computes computational complexity(FLOPs) of this Dectectron model?

youngwanLEE

i find this code only can use roialign ,if i use roipooling ,Memory-Usage of GPU increases when training until out of memory. i add some new roi such as psroi...

GYxiaOH