Caffe Installation on Ubuntu 14.04 log

Log

Caffe Installation
------------------

Versions

Cuda 6.5
Cudnn V1 for Cuda 6.5
Boost 1.55
OpenBLAS libopenblas-dev (0.2.8-6ubuntu1)
opencv-2.4.10
----------------------------
OpenBLAS

sudo apt-get install libopenblas-dev
----------------------------
Boost 1.55

Download and
sudo apt-get update
sudo apt-get install build-essential g++ python-dev autotools-dev libicu-dev build-essential libbz2-dev
./bootstrap.sh --prefix=/usr/local
sudo ./b2 --with=all -j 8 install
----------------------------
Cuda 6.5


sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev protobuf-compiler libhdf5-serial-dev
sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev

Download run file from cuda site
press ctrl-alt-F1
(become terminal tt1)
>> sudo /usr/bin/nvidia-uninstall

>> sudo service mdm stop
>> ./cuda_6.5.14_linux_64.run
-- follow instruction until installed complete
>> sudo service mdm start
 restart computer
>> sudo service mdm stop
>> ./cuda_6.5.14_linux_64.run
-- If you see "driver installed", then installation is complete


Open terminal
click on the middle of the terminal
goto profile>profile preference
choose title and command tab
select run command as login shell

vi ~/.profile
add to PATH /usr/local/cuda-6.5/bin
add to LD_LIBRARY_PATH /usr/local/cuda-6.5/lib64

----------------------------
cudnn V1.
Download, extract the tar file
sudo cp -r cudnn-6.5-linux-R1 /usr/local
add to PATH /usr/local/cudnn-6.5-linux-R1
add to LD_LIBRARY_PATH /usr/local/cudnn-6.5-linux-R1



----------------------INSTALL CAFFE-------------------

cp Makefile.config.example Makefile.config
make -j5 all
make -j5 test
make runtest
make pycaffe
make mat
----------------------INSTALL PYTHON PREREQUESITE----
sudo apt-get install numpy
sudo pip install matplotlib
sudo pip install numexpr
sudo pip install tables
sudo apt-get install scipy
=========END  =========
--------------------------- End Note ----------------------------------
Sometime, Cuda 6.5 may conflict with Neuveau

1. create file blacklist

$ vi /etc/modprobe.d/blacklist-nouveau.conf

blacklist nouveau
blacklist lbm-nouveau
options nouveau modeset=0
alias nouveau off
alias lbm-nouveau off

Then save

2. do this command 
echo options nouveau modeset=0 | sudo tee -a /etc/modprobe.d/nouveau-kms.conf

3. reboot

 
$ update-initramfs -u
$ reboot

4. Then Try to install Cuda 6.5 using above instruction again 

---------------- Makefile.config ---------------------
## Refer to http://caffe.berkeleyvision.org/installation.html
# Contributions simplifying and improving our build system are welcome!

# cuDNN acceleration switch (uncomment to build with cuDNN).
USE_CUDNN := 1

# CPU-only switch (uncomment to build without GPU support).
# CPU_ONLY := 1

# To customize your choice of compiler, uncomment and set the following.
# N.B. the default for Linux is g++ and the default for OSX is clang++
# CUSTOM_CXX := g++

# CUDA directory contains bin/ and lib/ directories that we need.
CUDA_DIR := /usr/local/cuda
# On Ubuntu 14.04, if cuda tools are installed via
# "sudo apt-get install nvidia-cuda-toolkit" then use this instead:
# CUDA_DIR := /usr

# CUDA architecture setting: going with all of them.
# For CUDA < 6.0, comment the *_50 lines for compatibility.
CUDA_ARCH := -gencode arch=compute_20,code=sm_20 \
        -gencode arch=compute_20,code=sm_21 \
        -gencode arch=compute_30,code=sm_30 \
        -gencode arch=compute_35,code=sm_35 \
        -gencode arch=compute_50,code=sm_50 \
        -gencode arch=compute_50,code=compute_50

# BLAS choice:
# atlas for ATLAS (default)
# mkl for MKL
# open for OpenBlas
BLAS := open
# Custom (MKL/ATLAS/OpenBLAS) include and lib directories.
# Leave commented to accept the defaults for your choice of BLAS
# (which should work)!
# BLAS_INCLUDE := /path/to/your/blas
# BLAS_LIB := /path/to/your/blas

# This is required only if you will compile the matlab interface.
# MATLAB directory should contain the mex binary in /bin.
 MATLAB_DIR := /usr/local/MATLAB/R2012b
# MATLAB_DIR := /Applications/MATLAB_R2012b.app

# NOTE: this is required only if you will compile the python interface.
# We need to be able to find Python.h and numpy/arrayobject.h.
PYTHON_INCLUDE := /usr/include/python2.7 \
        /usr/lib/python2.7/dist-packages/numpy/core/include
# Anaconda Python distribution is quite popular. Include path:
# PYTHON_INCLUDE := $(HOME)/anaconda/include \
        # $(HOME)/anaconda/include/python2.7 \
        # $(HOME)/anaconda/lib/python2.7/site-packages/numpy/core/include

# We need to be able to find libpythonX.X.so or .dylib.
PYTHON_LIB := /usr/lib
# PYTHON_LIB := $(HOME)/anaconda/lib

# Whatever else you find you need goes here.
INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include /usr/local/cuda-6.5/include /usr/local/cudnn-6.5-linux-R1
LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib /usr/local/cuda-6.5/lib64 /usr/local/cudnn-6.5-linux-R1

BUILD_DIR := build
DISTRIBUTE_DIR := distribute

# Uncomment for debugging. Does not work on OSX due to https://github.com/BVLC/caffe/issues/171
# DEBUG := 1

# The ID of the GPU that 'make runtest' will use to run unit tests.
TEST_GPUID := 0