Hand – to – hand teaching to build caffe and handwritten number recognition
\
\
Time: November 9, 2016 communication: Deep learning practical communication Q group 472899334, if you have questions, you can add this group to communicate together. To explore the principles behind experiments, see deep Learning Online. \
\
\
One, foreword
In the previous tutorial, we built TensorFlow and Torch, and after the tutorial was released, you had far fewer questions. However, caffe also has many problems, and caffe is one of the three frames to be covered in the deep Learning online class. Therefore, we went through caffe’s construction again, hand in hand and command prompt throughout. This tutorial is based on github: github.com/BVLC/caffe and page P28 of 21 Days caffe.
In addition, my installation environment is Ubuntu14.04, CUDA8.0, CudNn5.1, OpenCV, GTX1070. Check out the installation tutorials in “Flappy Bird” and “Van Gogh Painting with Tensorflow” for more information.
\
Install dependencies
Update the source
sudo apt-get update
Refer to the website page address: http://caffe.berkeleyvision.org/install_apt.html
Install command:
sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libhdf5-serial-dev protobuf-compiler
sudo apt-get install –no-install-recommends libboost-all-dev
Note: Holding CTRL + Shift + C on the Ubuntu command line is to copy, and CTRL + Shift + V is to paste
Iii. Related installation
Install git command:
sudo apt-get install git
Install the BLAS command:
sudo apt-get install libatlas-base-dev
Dependencies required to install the PyCaffe interface:
sudo apt-get install -y python-numpy python-scipy python-matplotlib python-sklearn python-skimage python-h5py python-protobuf python-leveldb python-networkx python-nose python-pandas python-gflags cython ipython
Install other dependencies:
sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev
4. Caffe related operations
Download caffe:
sudo git clone Github.com/BVLC/caffe….
Enter the caffe:
cd caffe
Configure the makefile. config file:
sudo cp Makefile.config.example Makefile.config
Modify the Makefile. Config:
sudo vi Makefile.config
I’m using Cudnn here, so I’ll go to the fourth line
# USE_CUDNN := 1
USE_CUDNN := 1
Schematic diagram:
Compile the caffe:
Run the following commands in sequence
sudo make all -j16
sudo make test -j16
sudo make runtest -j16
The caffe file used to compile Python
Compile the caffe:
Run the following command
make pycaffe -j16
Six, validation,
cd python
python
import caffe
No error indicates that the installation is successful!
\
\
Handwritten number recognition based on CAFFe
Caffe MNIST by Xiao CAI Github: github.com/BVLC/caffe CD /caffe/caffe 1. Download data./data/mnist/get_mnist.sh 2. Convert it to the LMDB format./examples/mnist/create_mnist.sh 3. Training data./examples/mnist/train_lenet.sh
\
On the other, caffe under MAC installation, please refer to: ask.julyedu.com/question/74… . July online ta team, November 9, 2016.