Caffe2
Precio de Caffe2: gratis
Versión: 0.8.1
Caffe2 - a platform for deep learning, sponsored by Facebook. The main advantages of Caffe2 are openness, speed and modularity.
The emphasis of Caffe2 is on mobile computing. The platform is optimized for ARM CPUs, supports iOS and Android.
The new features of the camera in Facebook Messenger to add effects to the images are based on Caffe2.
The Caffe2 platform can be used to program the capabilities of artificial intelligence on mobile devices - with help of Caffe2 you can recognize images, text, video, etc.
To write learning models, you only need a few lines of code. Subsequently, models can easily be turned into full-fledged applications.
Caffe2 is effectively used when creating chats.
Official site provides ready-made models that can be used in training of neural networks. As Nvidia® tests show, Caffe2 works much faster than its predecessor, Caffe.
Instructions for assembling Caffe2 from sources
1. Install
Cudnn. For our example, we used version 5.1:
CUDNN_URL="http://developer.download.nvidia®.com/compute/redist/cudnn/v5.1/cudnn-8.0-linux-x64-v5.1.tgz"
wget ${CUDNN_URL}
sudo tar -xzf cudnn-8.0-linux-x64-v5.1.tgz -C /usr/local
rm cudnn-8.0-linux-x64-v5.1.tgz && sudo ldconfig
2. Install
dependencies:
sudo apt-get update
sudo apt-get install -y --no-install-recommends \
build-essential \
cmake \
git \
libgoogle-glog-dev \
libgtest-dev \
libiomp-dev \
libleveldb-dev \
liblmdb-dev \
libopencv-dev \
libopenmpi-dev \
libsnappy-dev \
libprotobuf-dev \
openmpi-bin \
openmpi-doc \
protobuf-compiler \
python-dev \
python-pip
sudo pip install \
future \
numpy \
protobuf
# for Ubuntu 14.04
sudo apt-get install -y --no-install-recommends libgflags2
# for Ubuntu 16.04
sudo apt-get install -y --no-install-recommends libgflags-dev
3. Clone
code from GitHub:
git clone --recursive https://github.com/pytorch/pytorch.git && cd pytorch
git submodule update --init
4. Create a directory for Caffe2:
mkdir build && cd build
5. Configure the assembly.
cmake ..
If you encounter errors during installation at the "Linking CXX executable ../bin/mpi_test" stage, then run the command:
cmake .. -DUSE_MPI=OFF
6. Compile and install Caffe2:
sudo make install
7. Verify Caffe2.
cd ~ && python -c 'from caffe2.python import core' 2>/dev/null && echo "Success" || echo "Failure"
If Caffe2 was installed successfully, then you should see the message "Success".
Categoría
Machine learning
Industria
Automotive, Aerospace, Electronic, Biotechnology, Medicine, Finance
Licencia
Opensource/Free