Comments (7)
Hey, I solve this problem!
The problem lies in lib/build.sh. What I do is just to add '-gencode arch=compute_70,code=sm_70' (according to my GPU device RTX2070) to the CUDA_ARCH and recompile the NMS extension. Now everything works well!
Hope this help someone and sorry for my poor English.
from pytorch-retinanet.
Here's a bug I found in compiling nms.. hope it is useful to someone.
The build.sh file is saying "$CUDA_ARCH" as a parameter to nvcc command which was not set inside my docker environment.
The actual nvcc command goes like the one in the link below.
https://github.com/multimodallearning/pytorch-mask-rcnn
I replaced $CUDA_ARCH with --arch=sm_52 (I have a titanx).
I was getting a "no gpu_nms" module found error before and it was gone after compiling nms properly.
Hope this helps someone.
from pytorch-retinanet.
Hi,
I'm looking into this issue. Could you please indicate:
- Were you able to compile the NMS extension without any errors?
- Is this running in a jupyer notebook ?
Thanks.
from pytorch-retinanet.
Thanks for your reply.
- I think I compile the NMS extension without any errors:
root@9a42af0f41ef:/pytorch-retinanet/lib# sh build.sh
Compiling nms kernels by nvcc...
Including CUDA code.
/pytorch-retinanet/lib/nms
generating /tmp/tmpKGuPdx/_nms.c
setting the current directory to '/tmp/tmpKGuPdx'
running build_ext
building '_nms' extension
creating pytorch-retinanet
creating pytorch-retinanet/lib
creating pytorch-retinanet/lib/nms
creating pytorch-retinanet/lib/nms/src
x86_64-linux-gnu-gcc -pthread -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -fno-strict-aliasing -Wdate-time -D_FORTI
FY_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security -fPIC -DWITH_CUDA -I/usr/local/lib/python2.7/dist-packages/torch/utils/ffi/../../lib/include -I/usr/local/lib/python2.7/dist-packages/torch/utils/ffi/../../lib/include/TH -I/usr/local/lib/python2.7/dist-packages/torch/utils/ffi/../../lib/include/THC -I/usr/local/cuda/include -I/usr/include/python2.7 -c _nms.c -o ./_nms.o -std=c99 -std=c99x86_64-linux-gnu-gcc -pthread -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -fno-strict-aliasing -Wdate-time -D_FORTI
FY_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security -fPIC -DWITH_CUDA -I/usr/local/lib/python2.7/dist-packages/torch/utils/ffi/../../lib/include -I/usr/local/lib/python2.7/dist-packages/torch/utils/ffi/../../lib/include/TH -I/usr/local/lib/python2.7/dist-packages/torch/utils/ffi/../../lib/include/THC -I/usr/local/cuda/include -I/usr/include/python2.7 -c /pytorch-retinanet/lib/nms/src/nms.c -o ./pytorch-retinanet/lib/nms/src/nms.o -std=c99 -std=c99x86_64-linux-gnu-gcc -pthread -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -fno-strict-aliasing -Wdate-time -D_FORTI
FY_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security -fPIC -DWITH_CUDA -I/usr/local/lib/python2.7/dist-packages/torch/utils/ffi/../../lib/include -I/usr/local/lib/python2.7/dist-packages/torch/utils/ffi/../../lib/include/TH -I/usr/local/lib/python2.7/dist-packages/torch/utils/ffi/../../lib/include/THC -I/usr/local/cuda/include -I/usr/include/python2.7 -c /pytorch-retinanet/lib/nms/src/nms_cuda.c -o ./pytorch-retinanet/lib/nms/src/nms_cuda.o -std=c99 -std=c99/pytorch-retinanet/lib/nms/src/nms_cuda.c: In function ‘gpu_nms’:
/pytorch-retinanet/lib/nms/src/nms_cuda.c:29:35: warning: initialization from incompatible pointer type [-Wincompatible-p
ointer-types] unsigned long long* mask_flat = THCudaLongTensor_data(state, mask);
^
/pytorch-retinanet/lib/nms/src/nms_cuda.c:37:40: warning: initialization from incompatible pointer type [-Wincompatible-p
ointer-types] unsigned long long * mask_cpu_flat = THLongTensor_data(mask_cpu);
^
/pytorch-retinanet/lib/nms/src/nms_cuda.c:40:39: warning: initialization from incompatible pointer type [-Wincompatible-p
ointer-types] unsigned long long* remv_cpu_flat = THLongTensor_data(remv_cpu);
^
/pytorch-retinanet/lib/nms/src/nms_cuda.c:23:7: warning: unused variable ‘boxes_dim’ [-Wunused-variable]
int boxes_dim = THCudaTensor_size(state, boxes, 1);
^
x86_64-linux-gnu-gcc -pthread -shared -Wl,-O1 -Wl,-Bsymbolic-functions -Wl,-Bsymbolic-functions -Wl,-z,relro -fno-strict-
aliasing -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -Wdate-time -D_FORTIFY_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security -Wl,-Bsymbolic-functions -Wl,-z,relro -Wdate-time -D_FORTIFY_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security ./_nms.o ./pytorch-retinanet/lib/nms/src/nms.o ./pytorch-retinanet/lib/nms/src/nms_cuda.o /pytorch-retinanet/lib/nms/src/cuda/nms_kernel.cu.o -o ./_nms.so
- I run it in docker(on a Linux machine) instread of jupyter notebook.
Thanks.
from pytorch-retinanet.
@yhenon Hi have you figured out the problem of nms yet?I think I confront the same problem but I run it in a jupyter notebook. Thx
Well, I've read the codes about NMS and I think there is a problem as using only one pass of nms which treated all the detected objects in the same category. Shouldn't it using nms several times over each categories?
However, I'm not quite sure that this problem related to @CPFLAME yours
from pytorch-retinanet.
@qq184861643 HI
Agree with u.
In my experiment, although visualize works fine, there was no problem. But i think we should do nms once for each category when using NMS.
@yhenon
hope your help!!
many many thanks!!
from pytorch-retinanet.
I come across the same problem! The NMS extension has been compiled without any errors. I don't know how to solve it. Can anyone help me?
from pytorch-retinanet.
Related Issues (20)
- Loss curve
- How to specify GPU card number for training
- mAP Problem/question
- Question about NMS step in inference phase? HOT 1
- Question about input size HOT 1
- Possible discrepancy in FocalLoss calculation HOT 2
- Yolo format HOT 1
- multi-GPU training
- evaluate question
- Training Issue HOT 3
- Is batch testing possible? HOT 1
- Save module every time?
- Failure to switch to ONNX HOT 1
- IndexError: index -1 is out of bounds for axis 0 with size 0 HOT 3
- Early Stopping
- Alternative dataset
- i can't find the CSVGenerator,could you please tell me where is it? thank you
- Why is AP so low?(No changes were made. )
- Can we enhance retinanet model??
- Is there any training logs?
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from pytorch-retinanet.