profile
viewpoint
Sungjun Hong sungjunhong Sungkonghoe University Seoul, Korea Assistant Professor

sungjunhong/enjoynet 0

Convolutional Neural Networks in TensorFlow

sungjunhong/skhu.hclass 0

홍박의 느티아래 클라-쓰.

sungjunhong/YOLO-tensorflow 0

Tensorflow implementation of YOLO, including training and test phase.

yonsei-cilab/kros-2019-summer 0

한국로봇학회 제 4회 인공지능 및 로보틱스 여름학교 1일차 (8/19, 월) 오전 실습자료

push eventyankailab/OpenKAI

Kai Yan

commit sha a2cb50c35a4ee43cc8791efddac76507a3293e36

Testing and Updating AProver followTag

view details

push time in 6 hours

issue commentBVLC/caffe

how to install caffe into ubuntu 20.04?

Hi, Don't know if that is the case, but this post really helped me: https://medium.com/@accssharma/image-feature-extraction-using-pretrained-models-in-caffe-491d0c0b818b My advice to install caffe with gpu capability is: start by downloading the right cudda driver and then try to add to Makefile.config the includes that makefile eventually can't find. A small note if you're planning to work with TensorFlow: I've read some issues where installation problems were recurrent. while installing cuda 11, so if that eventually happens, consider installing cuda 10.

longzeyilang

comment created time in a day

push eventyankailab/OpenKAI

Kai Yan

commit sha 57db265ff99fb19147dd614666cc4ddf8e8f7efc

Testing AProver KU

view details

push time in a day

push eventyankailab/OpenKAI

Kai Yan

commit sha 7b428576fe2a97f3da693db8caf8f6c1a880c602

Updating apRover followTag

view details

push time in a day

push eventyankailab/OpenKAI

Kai Yan

commit sha 38def12c1f66d32cf9fe582fde29e56d4ff364de

Updatin apRover followTag

view details

push time in 2 days

issue commentBVLC/caffe

'../lib/libcaffe.so.1.0.0: undefined reference to `forkpty'' caffe build error, can anyone help me?

I meet this error too, have you solved it?

Qunxiang

comment created time in 3 days

issue commentBVLC/caffe

Unable to build with CUDA-10.2 + CuDNN-8.0.1

Thanks for your update.

On Thu, Nov 26, 2020 at 5:44 PM HYK-baby notifications@github.com wrote:

This problem puzzled me too. It token me 4 hours to make it work. Open the file “cmake/Cuda.cmake”. Find "file(READ ${CUDNN_INCLUDE}/cudnn.h CUDNN_VERSION_FILE_CONTENTS)",and replace "cudnn.h" with "cudnn_version.h".

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/BVLC/caffe/issues/6982#issuecomment-734190831, or unsubscribe https://github.com/notifications/unsubscribe-auth/AG4GR5BKFKS2SQROFDI7LYLSRYPPLANCNFSM4TDWOBEA .

MuhammadAsadJaved

comment created time in 4 days

issue commentBVLC/caffe

Unable to build with CUDA-10.2 + CuDNN-8.0.1

This problem puzzled me too. It token me 4 hours to make it work. Open the file “cmake/Cuda.cmake”. Find "file(READ ${CUDNN_INCLUDE}/cudnn.h CUDNN_VERSION_FILE_CONTENTS)",and replace "cudnn.h" with "cudnn_version.h".

MuhammadAsadJaved

comment created time in 4 days

startednaver/dcmm

started time in 6 days

push eventyankailab/OpenKAI

Kai Yan

commit sha b7e0b940255f1577d40547f6c90e3f1bb7557c61

Updating AProver

view details

push time in 6 days

issue commentBVLC/caffe

Segmentation fault in using python layer in caffe

i find when i use debug layer ,i can only use the first GPU ! dose Anybody know why

prateek-kacker

comment created time in 6 days

created repositorycarpedm20/test3

created time in 7 days

issue commentBVLC/caffe

This file was generated by an older version of protoc which is incompatible with your Protocol Buffer headers. Please regenerate this file with a newer version of protoc.

both were the same version. after that I used make clean, and then make test, and voila! the problem solved for my case.

@ghost

currently, the above two (source/apt) are the same version on my machine, but i still got this error.. Do you know

    1. how did you make and make clean, in which path?
    1. why i still have this problem...
SreehariRamMohan

comment created time in 7 days

GollumEvent

push eventyankailab/OpenKAI

Kai Yan

commit sha 7521142ee45c814ca98bf900dc61261770a7da4f

Updating AProver_follow

view details

push time in 7 days

PR opened BVLC/caffe

Fix typo

favorate -> favorite

+1 -1

0 comment

1 changed file

pr created time in 9 days

push eventyankailab/OpenKAI

Kai Yan

commit sha 482a56b09a50fcf3564c28b554148c1a8ec80ab4

Added Chilitags

view details

push time in 9 days

startedgraykode/commit-autosuggestions

started time in 10 days

startedgraykode/ai-docstring

started time in 10 days

issue openedBVLC/caffe

Lower top1/top5 accuracy on ImageNet for reference model

Issue summary

I tried to reproduce the accuracy values (top1 and top5) for the reference ImageNet model (CaffeNet). However, I only get a top5 accuracy of ~67% which is lower than the reported 80% here.

Steps to reproduce

I used the model definition and weights from GitHub. I evaluated the model using this simple script. I used the pre-processing (transforming from RGB to BGR, image resizing, center cropping, and ImageNet mean subtraction) indicated in the ImageNet examples of Caffe.

I uploaded the predicted classes and the ground-truth labels as a python list here, in case someone wants to take a look at them,

Tried solutions

I redownloaded the network weights and verified the preprocessing.

System configuration

  • Operating system: see below
  • Compiler:
  • CUDA version (if applicable):
  • CUDNN version (if applicable):
  • BLAS:
  • Python version (if using pycaffe):
  • MATLAB version (if using matcaffe):

I used the latest gpu docker container.

Issue checklist

  • [x] read the guidelines and removed the first paragraph
  • [x] written a short summary and detailed steps to reproduce
  • [x] explained how solutions to related problems failed (tick if found none)
  • [x] filled system configuration
  • [x] attached relevant logs/config files (tick if not applicable)

created time in 11 days

startedSeanNaren/deepspeech.pytorch

started time in 13 days

startedmozilla/DeepSpeech

started time in 13 days

issue commentBVLC/caffe

caffe make error while compiling

Did you find a fix/

mohamadmahdavian

comment created time in 13 days

startedseemoo-lab/opendrop

started time in 13 days

push eventyankailab/OpenKAI

Kai Yan

commit sha 44156bfd42c8221732f0da4c9a5c7e99c731b4d7

Updated PCsend

view details

push time in 14 days

push eventyankailab/OpenKAI

Kai Yan

commit sha 1bb70b0b97f7d3791b94ed5ed7985285d375a4cd

Updated PCsend and PCrecv

view details

push time in 15 days

startedmuhyun/autogluon_stack_visualizer

started time in 15 days

push eventyankailab/OpenKAI

Kai Yan

commit sha 9432159a88ee522d17e7e980c56efe6c6c34d620

Updated PCui and PCtransform

view details

push time in 15 days

push eventyankailab/OpenKAI

Kai Yan

commit sha bab6f2aacc197a046172488ad2b0049a79501f8a

Updated PCui

view details

push time in 17 days

issue openedBVLC/caffe

opencv_world library

I have built opencv4.4 with opencv_world flag turned on. So, I only have a large single .so file which is opencv_world.so . But building caffe seems to need individual .so files of opencv:

CXX src/caffe/layer.cpp CXX src/caffe/parallel.cpp AR -o .build_release/lib/libcaffe.a LD -o .build_release/lib/libcaffe.so.1.0.0 /usr/bin/ld: cannot find -lopencv_core /usr/bin/ld: cannot find -lopencv_highgui /usr/bin/ld: cannot find -lopencv_imgproc collect2: error: ld returned 1 exit status make: *** [Makefile:583: .build_release/lib/libcaffe.so.1.0.0] Error 1

Is there any workaround this or do I need to rebuild opencv?

And also I was facing several issues with cudnn accelation on for cudnn 8.0. I suppose there isnt a fix to this yet?

  • CUDA version: 11
  • CUDNN version : 8
  • OpenCV : 4.4.0

created time in 17 days

more