gosh i wish i could give you a hug right now. you're the champion! you covered what other youtubers (i won't name names) have missed and saved me from hours of aimlessly fixing things.
Thank you for your great tutorial, now I can sleep after learning this. Haha. However, allow me to add some solution, if your Cuda is 11.1 or above. In the first build, I encountered error "nvcc fatal : Unsupported GPU architecture 'compute_30'", after exploring on stackoverlow, it seems Cuda 11.0 above is not support compute_30, so, it should be change to compute_80. You can change it by opening make file on darknet folder, and change this line From this: ARCH= -gencode arch=compute_50,code=[sm_50,compute_50] \ -gencode arch=compute_52,code=[sm_52,compute_52] \ -gencode arch=compute_61,code=[sm_61,compute_61] to this: ARCH= -gencode arch=compute_80,code=sm_80 \ -gencode arch=compute_52,code=sm_52 \ -gencode arch=compute_61,code=sm_61 And then, before you build yolo_cpp_dll.vcxproj, make sure you already paste getopt.c dan getopt.h from C:\Users\user\darknet\darknet-master\3rdparty to C:\Users\user\darknet\darknet-master\src and the last thing, before you build darknet.sln, right click darknet, go to Project Properties → Configuration Properties → C/C++ → General → Additional Include Directories. Add the path to the folder containing 'getopt.h' from C:\Users\user\darknet\darknet-master\src\ Also don't remove compute 80 in your property, when you build the darknet in Vc 2019. It will resolved the issue, if you build it with Higher GPU and Cuda 11.1 above Remember, Always using VC 2019!
I am getting this error while building yolo_cpp_dll. MSB4019 The imported project "C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Microsoft\VC\v160\BuildCustomizations\CUDA 10.1.props" was not found. Confirm that the expression in the Import declaration "C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Microsoft\VC\v160\\BuildCustomizations\CUDA 10.1.props" is correct, and that the file exists on disk. I created darknet folder in D Directory. What to do now?????
@@nimetullahnecmettin2421 In English I deleted everything and even formatted my pc. I have solved the problem like this In Turkish ben her şeyi sildim hatta bilgisayara format atıp tekrardan kurdum anca bu şekilde oldu. Bir noktadan sonra insana daralma geliyor
@@sefadogan4795 Sağolun .aynen ya bu konfigürasyon baya bir zormuş ama bu hocanın eline sağlık çok basıt anlatmış öbürlerine göre. Ben yorumlardan çözümü boldum. O yüzden hemen çözüldü sayılır ama mAP grafıkte gözükmediği için bir daha çalıştırıyorum .
@@TheCodingBug Yeah, cudnn64_8.dll just worked fine but i had another error. Darknet.exe blocked by windows. If people have this problem, just delete darknet.exe and generate it again.
Hello, i succeeded on running yolo with webcam, but my stream resolution is stuck at 640 x 360, do you know how to change the resolution to 1920 x 1080? thank you for your help, the tutorial was really helpful to me
I already have done it. You can find two tutorials on the channel. One focuses the dataset creation while the other one shows how to train darknet YOLOv4 on that custom data.
help...problem building darknet in vs19.....used the same versions used in the video.........44 error's....e.g :Error calling a __host__ function("__floorf") from a __device__ function("stair_gradient_kernel") is not allowed ........and manny more........error were in the following files : 'activation_kernals.cu' , 'blas_kernals.cu' and 'CUDA 10.1.targets'.........and 4 warnings: "variable "MISH_THRESHOLD" was declared but never referenced" file: activation_kernals.cu "variable "out_index" was declared but never referenced" file:blas_kernals.cu "variable "step" was set but never used" file:blas_kernals.cu , "variable "stage_id" was declared but never referenced" file:blas_kernals.cu.
Good afternoon, I follow your instructions, but when building yolo_cpp_dll I have a problem nvcc fatal : Unsupported gpu architecture 'compute_30' and a list of errors with code MSB3721
@@potemkin8108 do you mean at the time of building opencv? If yes, you will have to build it again with computer 6.1. If opencv works fine as of now, there must be some other issue.
I came from your opencv video you posted few months ago to solve cv2 file not found and now here. I cannot find opencv_world444 file. I went directly to install then x64. I didn't go to build then install. As in the opencv video you created a different folder install. So where do I find this opencv_world444 file????
@@jagdishgoklani9712 Then the file should be there. But the version would be different. For example, instead of opencv_world444 it could be opencv_world451
Hello, with which command can I detect object on all photos in a particular folder in yolov4? I want it to be done one by one in all the photos one after the other.
Sir, why my command prompt showing (base) C:\>C:\darknet\darknet-master\build\darknet\x64>darknet.exe detector test cfg/coco.data cfg/yolov4.cfg yolov4.weights Access is denied. pls help me
...or, you can do all this with a single command: "vcpkg.exe install --triplet x64-windows-static cuda cudnn opencv-cuda darknet[cuda,cudnn,opencv-cuda]"
Hi i wanted you to ask this: My data is working in cmd fine (for gpu) but when i wanted the run it in pycharm it doesn't give any error and not drawing bounding boxes on my webcam or video or sth can u help me
Hi @Ezgi Topcu, I can help you with that, just let me see your code to know if you are in the right way. I have some script that will give what you want using dnn modules from Opencv. But my scripts can only work with CPU executing it in Pycharm. If you want gpu-based inference you will need to run your python script with Anaconda Prompt
Hi! is it possible to build darknet with opencv installed, but not opencv-world? because my opencv installation returns many errors when i try to build it with opencv_world
To my knowledge, on Windows, you do need openCV world.dll file. However, you can use precompiled file also (if you have installed the matching CUDA and cuDNN). By precompiled I mean, if someone else has this file and you are able to install the same version of CUDA and cuDNN on which this file was compiled, you can just copy this file and proceed with the build.
@@TheCodingBug do you have an idea why my opencv build fails only when i select opencv_world? I'm losing my mind and i really need it, i'm getting desperate. i have the same software versions as yours
So when should i tick the opencv build world on cmake? because when i tick it on the last process before generate the project count on the solution file reduced massively from 203 to 130, and after all the build process on visual studio i cant install the cv2 because its not detecting any cv2 module
@@TheCodingBug already ticking it on the first configure with other ticking, projects count still on 130. decided to still go with it, installing on vs take very long time. but after finished and when try to import it still say Traceback (most recent call last): File "", line 1, in ModuleNotFoundError: No module named 'cv2'
@@muhammadrezafairuzi7328 Make sure you only have one version of PYthon, Anaconda based preferably ... if there are multiple versions, this problem can occur. CMake shows the paths to the python executables in the configuration panel where you tick options. You can verify if it has picked the right version.
I got this error during building the yolo_cpp_dll Severity Code Description Project File Line Suppression State Error MSB4019 The imported project "C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Microsoft\VC\v160\BuildCustomizations\CUDA 10.1.props" was not found. Confirm that the expression in the Import declaration "C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Microsoft\VC\v160\\BuildCustomizations\CUDA 10.1.props" is correct, and that the file exists on disk. yolo_cpp_dll C:\darknet\darknet-master\build\darknet\yolo_cpp_dll.vcxproj 55 Is there any solution?
Hi, I found a solution from internet Just copy all files from this path (depends on the path you installed CUDA in) C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.1\extras\visual_studio_integration\MSBuildExtensions to this path: C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Microsoft\VC\v160\BuildCustomizations Thank you for the reply
You have to build opencv and enable opencv world while configuring it in CMAKE. Or you can use this updated tutorial: th-cam.com/video/YsmhKar8oOc/w-d-xo.html
@@boringchannel6313 If you enabled opencv_world during CMake configuration, you should have opencv_world dll file. If its not there, you have missed something. Both versions i.e 4.4.0 and 4.5.1 work equally fine. If you do not have the OpenCV world dll file, I suggest building again but use the updated tutorial for OpenCV version 4.5.1 instead of version 4.4.0.
Hi again, i can use yolov4 and darknet perfectly after your tutorial. I did everything what you did, but i can not use python anymore. Libraries which use opencv are not working anymore. I cant import cv2 for example. It gives that error; file "", line 1, in modulenotfounderror: no module named 'cv2'. What should i do? I cant use both sypder and darknet together :(
@@canozkan9104 Once you get Darknet working, you do not need OpenCV older version. You can follow this tutorial to install OpenCV GPU to your Anaconda base or virtual environment. th-cam.com/video/YsmhKar8oOc/w-d-xo.html
I did it already before your object detection video. I checked again. Opencv only works in opencv_gpu env, which i created while was following your instructions. I have another question maybe you know answer. How can i edit label font size in predicted image? Bounding box is good but label's letters are too big.
@@HomonculusPort if you've built yolov4 with GPU/CUDA support, it should utilize GPU. Alternatively, you can use OpenCv version of YOLO. th-cam.com/video/FjyF03uawsA/w-d-xo.html
(base) C:\Users\user\Downloads>python Python 3.10.9 | packaged by Anaconda, Inc. | (main, Mar 1 2023, 18:18:15) [MSC v.1916 64 bit (AMD64)] on win32 Type "help", "copyright", "credits" or "license" for more information. >>> import cv2 as cv Traceback (most recent call last): File "", line 1, in File "C:\Users\user\anaconda3\lib\site-packages\cv2\__init__.py", line 181, in bootstrap() File "C:\Users\user\anaconda3\lib\site-packages\cv2\__init__.py", line 153, in bootstrap native_module = importlib.import_module("cv2") File "C:\Users\user\anaconda3\lib\importlib\__init__.py", line 126, in import_module return _bootstrap._gcd_import(name[level:], package, level) ImportError: DLL load failed while importing cv2: The specified module could not be found. I got this error when importing cv2. Does anyone know how to fix it?
Hi@TheCodingBug I have followed your tutorial since starting my Project with Yolov4(Darknet). I spent nearly four days to figure out my mistake to install opencv and finally, I made it with the help of your video(without tears). But now I am facing the following error with visual studio 2022. Once I open the file yolo_cpp_dll in visual studio 2022 I am facing this Error: Designtime build error for the project "C:\darknet\master\build\darknet\yolo_cpp_dll.vcxproj", Release|x64 configuration. IntelliSense may not be available. Check the files "*.designtime.log" in your %TEMP% directory. For four days I have read many blogs but until now my progress is zero. I am not able to go ahead. I am trying my level best but losing hope and time day by day. can you please help me here by giving me some tips? your reply would save me. I am doing my career transition. so this is the project which is going to be a game-changer in my life. so please give me some tips or suggestions once you have time. I will expect your reply every second. Thank you.
![ทุกครั้งที่หลับตา (Lucid Dream) - AYLA's [ Official MV ]](http://i.ytimg.com/vi/4i1OBAQyv58/mqdefault.jpg)
Building OpenCV: th-cam.com/video/tjXkW0-4gME/w-d-xo.html
gosh i wish i could give you a hug right now. you're the champion! you covered what other youtubers (i won't name names) have missed and saved me from hours of aimlessly fixing things.
I am glad I was able to help.
Excellent tutorials.. It's fast however very much understandable.... I like to see more projects from you..
Thank you for your appreciation and the suggestion. I will try to slow down the pace in the next videos.
Thank you for your great tutorial, now I can sleep after learning this. Haha.
However, allow me to add some solution, if your Cuda is 11.1 or above. In the first build, I encountered error "nvcc fatal : Unsupported GPU architecture 'compute_30'", after exploring on stackoverlow, it seems Cuda 11.0 above is not support compute_30, so, it should be change to compute_80. You can change it by opening make file on darknet folder, and change this line
From this:
ARCH= -gencode arch=compute_50,code=[sm_50,compute_50] \
-gencode arch=compute_52,code=[sm_52,compute_52] \
-gencode arch=compute_61,code=[sm_61,compute_61]
to this:
ARCH= -gencode arch=compute_80,code=sm_80 \
-gencode arch=compute_52,code=sm_52 \
-gencode arch=compute_61,code=sm_61
And then, before you build yolo_cpp_dll.vcxproj, make sure you already paste getopt.c dan getopt.h from C:\Users\user\darknet\darknet-master\3rdparty to C:\Users\user\darknet\darknet-master\src
and the last thing, before you build darknet.sln, right click darknet, go to Project Properties → Configuration Properties → C/C++ → General → Additional Include Directories.
Add the path to the folder containing 'getopt.h' from C:\Users\user\darknet\darknet-master\src\
Also don't remove compute 80 in your property, when you build the darknet in Vc 2019.
It will resolved the issue, if you build it with Higher GPU and Cuda 11.1 above
Remember, Always using VC 2019!
Hello thanks for great tutorial! I've made it to the end, but how can I build my custom train with the darknet?
Look forward hearing from you soon!
th-cam.com/video/sKDysNtnhJ4/w-d-xo.html
Thank you so much for your awesome explanation!
I am glad you found it helpful.
This video was very helpful, it worked for me, thanks a lot. Vietnamese with love
I am glad I could help.
@@TheCodingBug Do you have video tutorials on building the darknet and debugging?
I am getting this error while building yolo_cpp_dll. MSB4019 The imported project "C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Microsoft\VC\v160\BuildCustomizations\CUDA 10.1.props" was not found. Confirm that the expression in the Import declaration "C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Microsoft\VC\v160\\BuildCustomizations\CUDA 10.1.props" is correct, and that the file exists on disk.
I created darknet folder in D Directory. What to do now?????
Did you fix this error? I can't
@@sefadogan4795 I have the same error.
I have also the same error. Is there any who can solve this problem?
@@nimetullahnecmettin2421
In English
I deleted everything and even formatted my pc. I have solved the problem like this
In Turkish
ben her şeyi sildim hatta bilgisayara format atıp tekrardan kurdum anca bu şekilde oldu. Bir noktadan sonra insana daralma geliyor
@@sefadogan4795 Sağolun .aynen ya bu konfigürasyon baya bir zormuş ama bu hocanın eline sağlık çok basıt anlatmış öbürlerine göre. Ben yorumlardan çözümü boldum. O yüzden hemen çözüldü sayılır ama mAP grafıkte gözükmediği için bir daha çalıştırıyorum .
Nice Tutorial must watch it.
Thank You.
I am getting comute30 and compute70 errors sir, please help how to fix it
i have cuda version 11.2, and when i build in visual studio, it says error about the SDK, do you know why?
even i have the same problem!
Last version of cudnn doesn't have cudnn64_7.dll, what should we do about that ?
Do you mean the latest version?
@@TheCodingBug Yes latest version of cudnn which is cudnn v8.2.1, there is a dll named cudnn64_8.dll but not cudnn64_7.dll
@@canozkan9104 then you'll need to compile yolov4 with new versions of cuda and cuDNN. And your file name would be different based on your version.
@@TheCodingBug Yeah, cudnn64_8.dll just worked fine but i had another error. Darknet.exe blocked by windows. If people have this problem, just delete darknet.exe and generate it again.
Hello, i succeeded on running yolo with webcam, but my stream resolution is stuck at 640 x 360, do you know how to change the resolution to 1920 x 1080? thank you for your help, the tutorial was really helpful to me
Superb job.. so impressive ..
Thanks for the appreciation.
I don't have file cudnn64_7 but cudnn64_8 is it okay?
I also have cudnn64_8. I think it is because of the new version of CUDA.
In build folder, opencv_world440.dll file is missing in my build folder. how can i generate it?
I want to know how to do darknet custom data training in Windows environment.
If you can afford it, can you make a tutorial?
Thank you.
It's best.
I already have done it. You can find two tutorials on the channel. One focuses the dataset creation while the other one shows how to train darknet YOLOv4 on that custom data.
Part 1: th-cam.com/video/sKDysNtnhJ4/w-d-xo.html
Part 2: th-cam.com/video/-NEB5P-SLi0/w-d-xo.html
Wow. Thank you very much. I have to try it. Thanks again.
I get Video-stream stopped error when i run it on videos, what could be the issue?
I can run it on images and web-cam, but for videos i am unable to
It seems there is an error with the video. Kindly try some other video, preferably .mp4 format.
@@TheCodingBug I tried with .mp4 format only, tried with several videos, still same issue
@@riteshananth6679 yess, same here.....did u got any solution?
any solutions??
help...problem building darknet in vs19.....used the same versions used in the video.........44 error's....e.g :Error calling a __host__ function("__floorf") from a __device__ function("stair_gradient_kernel") is not allowed ........and manny more........error were in the following files : 'activation_kernals.cu' , 'blas_kernals.cu' and 'CUDA 10.1.targets'.........and 4 warnings:
"variable "MISH_THRESHOLD" was declared but never referenced" file: activation_kernals.cu
"variable "out_index" was declared but never referenced" file:blas_kernals.cu
"variable "step" was set but never used" file:blas_kernals.cu ,
"variable "stage_id" was declared but never referenced" file:blas_kernals.cu.
I have following configuration i.e.
windows 10
GPU - NVIDIA 1660ti
Cuda 10.1 cudnn 7.5.1
can I install opencv 4.5.2 for yolo v3 and yolo v4?
I haven't tried this configuration. So I am not sure. Maybe someone else can help here.
Good afternoon, I follow your instructions, but when building yolo_cpp_dll I have a problem
nvcc fatal : Unsupported gpu architecture 'compute_30'
and a list of errors with code MSB3721
It seems you have old GPU.
@@TheCodingBug i have 1070...
@@potemkin8108 your compute version should be 6.1 not 3.0!!
@@TheCodingBug how i can change it? When builded i had choose 6.1 in cmake
@@potemkin8108 do you mean at the time of building opencv?
If yes, you will have to build it again with computer 6.1.
If opencv works fine as of now, there must be some other issue.
I came from your opencv video you posted few months ago to solve cv2 file not found and now here. I cannot find opencv_world444 file. I went directly to install then x64. I didn't go to build then install. As in the opencv video you created a different folder install. So where do I find this opencv_world444 file????
You have to enable opencv_world while building opencv in Cmake.
@@TheCodingBug So do I need to do the whole process of opencv again???
Because in that opencv video we enabled build_opencv_world....
@@jagdishgoklani9712 Then the file should be there. But the version would be different. For example, instead of opencv_world444 it could be opencv_world451
Do you know how to run a single darknet instance for multiple cameras?
In last step I want build into darknet.exe in Visual studio,it would fail,and than it must save as Unicode ,but what should I do?
in the previous video "opencv dnn module" i have successfully saved video output. how to saved video output in this tutorial?
Hello, with which command can I detect object on all photos in a particular folder in yolov4? I want it to be done one by one in all the photos one after the other.
Really Cool :)
Its still work on 2023?
Sir, why my command prompt showing (base) C:\>C:\darknet\darknet-master\build\darknet\x64>darknet.exe detector test cfg/coco.data cfg/yolov4.cfg yolov4.weights
Access is denied. pls help me
...or, you can do all this with a single command: "vcpkg.exe install --triplet x64-windows-static cuda cudnn opencv-cuda darknet[cuda,cudnn,opencv-cuda]"
Thank you for sharing.
Do you do this after installing visual Studio? Can you explain in detail?
@@feyzatopcu1342 see: github.com/microsoft/vcpkg
Is it required to have a nvidia gpu to run yolov4?
Hi I am getting problem while downloading Cudnn file so can send it.
Can you do real time webcam detection using yolov4 or yolov3? tq
Hi. Yes you can. It is explained at the end of the video. You can check the command in description as well.
@@TheCodingBug Correction*, web app that can perform real time webcam detection
Hi i wanted you to ask this: My data is working in cmd fine (for gpu) but when i wanted the run it in pycharm it doesn't give any error and not drawing bounding boxes on my webcam or video or sth can u help me
Hi @Ezgi Topcu, I can help you with that, just let me see your code to know if you are in the right way. I have some script that will give what you want using dnn modules from Opencv. But my scripts can only work with CPU executing it in Pycharm. If you want gpu-based inference you will need to run your python script with Anaconda Prompt
@@andrespereira4852 I really thank you i solved it :)
Hi! is it possible to build darknet with opencv installed, but not opencv-world? because my opencv installation returns many errors when i try to build it with opencv_world
To my knowledge, on Windows, you do need openCV world.dll file. However, you can use precompiled file also (if you have installed the matching CUDA and cuDNN). By precompiled I mean, if someone else has this file and you are able to install the same version of CUDA and cuDNN on which this file was compiled, you can just copy this file and proceed with the build.
@@TheCodingBug do you have an idea why my opencv build fails only when i select opencv_world? I'm losing my mind and i really need it, i'm getting desperate. i have the same software versions as yours
@@anto__729 what's the error?
BTW, you can try alternative method th-cam.com/video/YsmhKar8oOc/w-d-xo.html
@@TheCodingBug It gives me 2 unresolved external errors on the following file: convolution_layer.obj
The other method gives the sam errors :(
So when should i tick the opencv build world on cmake? because when i tick it on the last process before generate the project count on the solution file reduced massively from 203 to 130, and after all the build process on visual studio i cant install the cv2 because its not detecting any cv2 module
Do it for the first time when you start ticking the options.
@@TheCodingBug so before the other 4 option? The cuda, dnn, and math thing
@@TheCodingBug already ticking it on the first configure with other ticking, projects count still on 130. decided to still go with it, installing on vs take very long time. but after finished and when try to import it still say
Traceback (most recent call last):
File "", line 1, in
ModuleNotFoundError: No module named 'cv2'
@@muhammadrezafairuzi7328 Make sure you only have one version of PYthon, Anaconda based preferably ... if there are multiple versions, this problem can occur. CMake shows the paths to the python executables in the configuration panel where you tick options. You can verify if it has picked the right version.
@@TheCodingBug Already checked on cmake, the python 3 executable is already correct with the directory goes to anaconda. Anymore ideas?
Hi is there any way to get Yolov4 worknig on an RTX 3000 series? The problem is that the cards require CUDA 11 and cuDnn 8....
Yes you can use CUDA 11 and cuDNN 8 instead of older version.
@@TheCodingBug but what files do we need to copy from cuDNN and OpenCV !
@@G_B_R_N same files that I've copied. But their versions would be different.
I got this error during building the yolo_cpp_dll
Severity Code Description Project File Line Suppression State
Error MSB4019 The imported project "C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Microsoft\VC\v160\BuildCustomizations\CUDA 10.1.props" was not found. Confirm that the expression in the Import declaration "C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Microsoft\VC\v160\\BuildCustomizations\CUDA 10.1.props" is correct, and that the file exists on disk. yolo_cpp_dll C:\darknet\darknet-master\build\darknet\yolo_cpp_dll.vcxproj 55
Is there any solution?
Make sure you have installed CUDA 10.1 and same is mentioned in the two files that I edit.
@@TheCodingBug I've already installed CUDA 10.1 and changed the CUDA version in that file to CUDA 10.1 but still doesn't work
@@rickyhartono7140 I never encountered this error. I am afraid I cannot be of any help here.
Hi, I found a solution from internet
Just copy all files from this path (depends on the path you installed CUDA in)
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.1\extras\visual_studio_integration\MSBuildExtensions
to this path:
C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Microsoft\VC\v160\BuildCustomizations
Thank you for the reply
@@rickyhartono7140 Thank you for sharing. I hope it helps someone else!
I am not able tu download cuDNN it's required Nvidia program membership. How can I get those file without membership 🥵🥵
4:37 i dont have something called opencv_world440.dll :( can u help me?
You have to build opencv and enable opencv world while configuring it in CMAKE. Or you can use this updated tutorial: th-cam.com/video/YsmhKar8oOc/w-d-xo.html
@@TheCodingBug I already built it, should i do it again or i can continue with old buildings?
@@boringchannel6313 If you enabled opencv_world during CMake configuration, you should have opencv_world dll file. If its not there, you have missed something.
Both versions i.e 4.4.0 and 4.5.1 work equally fine. If you do not have the OpenCV world dll file, I suggest building again but use the updated tutorial for OpenCV version 4.5.1 instead of version 4.4.0.
When cmake, python3 path goes to default python folder, not to anaconda3. How to change this.
You change paths in cmake by clicking on them.
@@TheCodingBug Thank you, i am using cuda 11.4 with rtx2060 maxq ver, there is no cudnn64_7.dll file in cuda bin folder.
@@visakr6101 you have to download cuDNN separately. You can follow the tutorial to install CUDA and cuDNN from the link in the description.
I can’t find opencv_world440 inside the bin folder
You need to enable it during building OpenCV. Check this updated tutorial: th-cam.com/video/YsmhKar8oOc/w-d-xo.html
what if i have latest Cuda and Cudnn versions installed. Can we downgrade them ?
i do not think so, even it is possible.. it may be very straightforward.
I mean NOT straightforward...
Sir while installing cuda 7zip extraction error is occurring
Use WinRAR instead.
This works with AMD GPU?
No it doesn't utilize AMD GPU.
Hi, you can help me please
I do not have nvidea GPU, somebody help!
then try using colab gpu: th-cam.com/video/mmj3nxGT2YQ/w-d-xo.html
Hi again, i can use yolov4 and darknet perfectly after your tutorial. I did everything what you did, but i can not use python anymore. Libraries which use opencv are not working anymore. I cant import cv2 for example. It gives that error; file "", line 1, in modulenotfounderror: no module named 'cv2'. What should i do? I cant use both sypder and darknet together :(
Can you use OpenCV in Anaconda prompt?
@@TheCodingBug Unfortunately no
@@canozkan9104 Once you get Darknet working, you do not need OpenCV older version. You can follow this tutorial to install OpenCV GPU to your Anaconda base or virtual environment.
th-cam.com/video/YsmhKar8oOc/w-d-xo.html
I did it already before your object detection video. I checked again. Opencv only works in opencv_gpu env, which i created while was following your instructions. I have another question maybe you know answer. How can i edit label font size in predicted image? Bounding box is good but label's letters are too big.
@@TheCodingBug can you provide your mail id I have some doubt
Is it possible to speed up the detection speed from 13.5 FPS
Yes. We can use any light weight object detection model or yolo tiny versions. But for YOLO 4, it is maximum speed I could achieve.
TheCodingBug Thankyou for replying. When I ran YOLOv4 only maximum 15% of my gpu was being used. Do you know any method to increase that % ?
can i get all codes related to yolo Please
how to increase fps in video
it seems like its not utilizing my gpu its just 4fps on image vid and webcam
@@HomonculusPort if you've built yolov4 with GPU/CUDA support, it should utilize GPU.
Alternatively, you can use OpenCv version of YOLO.
th-cam.com/video/FjyF03uawsA/w-d-xo.html
@@TheCodingBug i think i did something wrong on the cmake or is it on the building part on visual studio?
@@HomonculusPort it'd be in cmake or while building opencv with GPU support.
Visual studio just builds the configurations.
(base) C:\Users\user\Downloads>python
Python 3.10.9 | packaged by Anaconda, Inc. | (main, Mar 1 2023, 18:18:15) [MSC v.1916 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import cv2 as cv
Traceback (most recent call last):
File "", line 1, in
File "C:\Users\user\anaconda3\lib\site-packages\cv2\__init__.py", line 181, in
bootstrap()
File "C:\Users\user\anaconda3\lib\site-packages\cv2\__init__.py", line 153, in bootstrap
native_module = importlib.import_module("cv2")
File "C:\Users\user\anaconda3\lib\importlib\__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
ImportError: DLL load failed while importing cv2: The specified module could not be found.
I got this error when importing cv2. Does anyone know how to fix it?
Hi@TheCodingBug I have followed your tutorial since starting my Project with Yolov4(Darknet). I spent nearly four days to figure out my mistake to install opencv and finally, I made it with the help of your video(without tears). But now I am facing the following error with visual studio 2022.
Once I open the file yolo_cpp_dll in visual studio 2022 I am facing this
Error: Designtime build error for the project "C:\darknet\master\build\darknet\yolo_cpp_dll.vcxproj", Release|x64 configuration. IntelliSense may not be available. Check the files "*.designtime.log" in your %TEMP% directory.
For four days I have read many blogs but until now my progress is zero. I am not able to go ahead. I am trying my level best but losing hope and time day by day. can you please help me here by giving me some tips? your reply would save me. I am doing my career transition. so this is the project which is going to be a game-changer in my life. so please give me some tips or suggestions once you have time. I will expect your reply every second. Thank you.