Git Product home page Git Product logo

decaf-release's Introduction

Decaf

Decaf is a framework that implements convolutional neural networks, with the goal of being efficient and flexible. It allows one to easily construct a network in the form of an arbitrary Directed Acyclic Graph (DAG) and to perform end-to-end training.

For more usage check out the wiki. A great place to start is running ImageNet classification on an image.

For the pre-trained imagenet DeCAF feature and its analysis, please see our technical report on arXiv. Please consider citing our paper if you use Decaf in your research:

@article{donahue2013decaf,
  title={DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition},
  author={Donahue, Jeff and Jia, Yangqing and Vinyals, Oriol and Hoffman, Judy and Zhang, Ning and Tzeng, Eric and Darrell, Trevor},
  journal={arXiv preprint arXiv:1310.1531},
  year={2013}
}

For Anaconda users experiencing libm error: it is because anaconda ships with a libm.so binary that does not support GLIBC_2.15, which gets loaded earlier than the system libm. You can fix this error by you can replacing anaconda's libm file with a newer version.

decaf-release's People

Contributors

dnouri avatar joshrosen avatar sergeyk avatar yangqing avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

decaf-release's Issues

Issue with MPI

Hi,

When I try to import the imagenet module, I get the following error:
screenshot from 2014-03-11 17 46 33

Could you tell me what is the issue here?

Can you provide the pre-trained ImageNet models?

Hi,
There is a network system still rely on DeCAF to work, including the imagenet.decafnet.epoch90 and the imagenet.decafnet.meta.
Can you provide these pre-trained ImageNet models for me? Thanks!

Can you provide the pre-trained ImageNet models

Hi,
There is a network system still rely on DeCAF to work, including the imagenet.decafnet.epoch90 and the imagenet.decafnet.meta.
Can you provide these pre-trained ImageNet models for me? Thanks!

unrecognized option -Wl

Hi, when I run python setup.py build on both my mac and linux machines I get an error while compiling: unrecognized option: "-Wl"

I edited the file layers/cpp/Makefile and removed the -Wl from the linking options and it built and installed just fine. I am not sure what that option does but that is how I made it work for me.

Thank you!

Getting an error: unrecognized command line option ‘-Wl’

Running python setup.py build I see errors about the -Wl flag

According to http://linux.die.net/man/1/g++ , -Wl is used to pass options - it looks like here there are no options in the Makefile in layers/cpp? Is there any reason for that flag to be there?

Error output

$python setup.py build
make -C layers/cpp/
make[1]: Entering directory `/home/kkastner/proj/decaf-release/decaf/layers/cpp'
g++ -c -fPIC -O3 -Wall -ffast-math -msse -msse2 -fopenmp im2col.cpp fastpool.cpp local_response_normalization.cpp neuron.cpp
g++ -shared -Wl -fopenmp -lgomp -o libcpputil.so *.o
g++: error: unrecognized command line option ‘-Wl’
make[1]: *** [all] Error 1
make[1]: Leaving directory `/home/kkastner/proj/decaf-release/decaf/layers/cpp'
make: *** [all] Error 2
running build
running build_py

Building on OSX Mavericks

To get decaf to build on OSX 10.9 (Mavericks) I had to modify decaf/layers/cpp/Makefile and remove the line "CC=g++" because 10.9 ships with clang installed by default, the current version of which doesn't support OpenMP.

Instead, I installed gcc from MacPorts, commented out that line, and did an "export CC=g++-mp-4.8" before running python setup.py build.

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo 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.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.