Comments (12)
Hello,
CNTK currently supports 7.0 only but I think it should be easy to modify it to compile with 7.5
Try doing the following: open Source\Math\MathCUDA.vcxproj and change following lines:
- CUDA_PATH_V7_0 to CUDA_PATH_V7_5
-
$(VCTargetsPath)\BuildCustomizations\CUDA 7.0.props to $ (VCTargetsPath)\BuildCustomizations\CUDA 7.5.props -
$(VCTargetsPath)\BuildCustomizations\CUDA 7.0.targets to $ (VCTargetsPath)\BuildCustomizations\CUDA 7.5.targets
Hope this helps, let me know if you still have problems compiling the code.
from cntk.
Many thanks, Alexey,
I got another problem. The EvaDll project cannot be loaded due to an error, "$(UsesCuda) evaluates to "" instead of a boolean". Do you have any idea about it?
Thanks,
from cntk.
Hmm, I don't know what's the problem yet, however, if everything else builds fine, you can still run CNTK even without EvalDll (which is really just a sample demonstrating how to load trained models). I know it's not ideal but hopefully will unblock you for now.
from cntk.
I built CNTK with CUDA 7.5. Here are the changes I made:
In ActionsLib.vcxproj, MathCUDA.vcxproj, and Math.vcxproj, there are a number of references to CUDA dll's, e.g. "cudart64_70.dll". Change the "70" version suffix to "75" , e.g. "cudart64_75.dll'
In MathCUDA.vcsproj, change the CudaPath attribute from "$(CUDA_PATH_V7_0)" to "$(CUDA_PATH_V7_5)". There is a reference to "CUDA 7.0.props" - change this to "CUDA 7.5.props". Similarly, change "CUDA 7.0.targets" to "CUDA 7.5.targets".
You can download a copy of my files here: https://www.dropbox.com/sh/7cbuxz2rde9gjxe/AADPaPS5UtovRuQFbBfnb3W7a?dl=0
from cntk.
Great! Thanks a lot. Actually I haven't gotten a successful build yet. I have another mess - both vs2012 and vs2013 installed on my machine, where I also run theano. Not sure if the issue of having two compilers has been addressed.
from cntk.
I have VS 2013 and 2015 on my machine. I think I did the build with 2013? What problem are you having with your build?
On Feb 12, 2016, at 5:50 PM, holmes2016 [email protected] wrote:
Great! Thanks a lot. Actually I haven't gotten a successful build yet. I have another mess - both vs2012 and vs2013 installed on my machine, where I also run theano. Not sure if the issue of having two compilers has been addressed.
—
Reply to this email directly or view it on GitHub.
from cntk.
After applied the changes made in your project files, I still got the following when build math project:
Error 1 error MSB4062: The "Nvda.Build.CudaTasks.SanitizePaths" task could not be loaded from the assembly C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V120\BuildCustomizations\Nvda.Build.CudaTasks.v7.5.dll. Could not load file or assembly 'Microsoft.Build.Utilities.v3.5, Version=3.5.0.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a' or one of its dependencies. The system cannot find the file specified. Confirm that the declaration is correct, that the assembly and all its dependencies are available, and that the task contains a public class that implements Microsoft.Build.Framework.ITask. C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V120\BuildCustomizations\CUDA 7.5.targets 160 9 MathCUDA
Error 2 error C2440: '' : cannot convert from 'initializer-list' to 'std::array<size_t,0x02>' (TensorView.cpp) c:\users\jlu\projects\cntk\source\math\TensorView.h 32 1 Math
from cntk.
Sorry, no idea what's going on there.
On Wed, Feb 17, 2016 at 9:12 AM, holmes2016 [email protected]
wrote:
After applied the changes made in your project files, I still got the
following when build math project:
Error 1 error MSB4062: The "Nvda.Build.CudaTasks.SanitizePaths" task could
not be loaded from the assembly C:\Program Files
(x86)\MSBuild\Microsoft.Cpp\v4.0\V120\BuildCustomizations\Nvda.Build.CudaTasks.v7.5.dll.
Could not load file or assembly 'Microsoft.Build.Utilities.v3.5,
Version=3.5.0.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a' or one
of its dependencies. The system cannot find the file specified. Confirm
that the declaration is correct, that the assembly and all its dependencies
are available, and that the task contains a public class that implements
Microsoft.Build.Framework.ITask. C:\Program Files
(x86)\MSBuild\Microsoft.Cpp\v4.0\V120\BuildCustomizations\CUDA 7.5.targets
160 9 MathCUDA
Error 2 error C2440: '' : cannot convert from 'initializer-list' to
'std::array' (TensorView.cpp)
c:\users\jlu\projects\cntk\source\math\TensorView.h 32 1 Math—
Reply to this email directly or view it on GitHub
#61 (comment).
from cntk.
No problem. I guess it is related to my VS and cuda setup. But my theano is able to run with cuda 7.5.
Thanks
from cntk.
We will be adding support for CUDA 7.5 within the next month.
from cntk.
Cool! Thanks
from cntk.
It seems like CUDA 7.5 support was added as of April 5th 2016.
https://github.com/Microsoft/CNTK/wiki/News#april-2016
from cntk.
Related Issues (20)
- Value goes invalid when using TestMinibatch
- Request for a no-opencv dotnet release
- CNTK C# Crash when layer is deeper
- Convolution 1D CNTK C++ HOT 1
- Add support to release linux aarch64 wheels
- Error while deploying MS Teams Bot with SSO
- how to install in python3.8
- A model causes CNTK crash with cudnnSetPoolingNDDescriptor when invoking the pooling operator
- This repo is missing important files HOT 1
- API document issue due to syntax in source code HOT 1
- Error: could not find all specified 'to_nodes' in clone. Looking for ['relu5'], found [None]
- SequenceClassification.py does not work out of box HOT 1
- Microsoft has absolutely no potential
- development of a neural network for object search
- join vcpkg
- cntk crash and pycharm process finished with exit code -1066598274 (0xC06D007E)
- ConvolutionTranspose2D outputs normally when num_filters is tuple and dilation is 0
- program crash when get gradient of `ConvolutionTranspose2D`
- when shape contains negative integer, input_variable should throw an error HOT 1
- MAX_POOLING crash, when pooling_window_shape contains 0 HOT 1
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 cntk.