從開學到現在,caffe裝了有4-5次了。在這里做個總結,以防那天,自己的電腦又操作失誤,又跪!
建議,如果是自己的電腦,能用網線,可以這樣搞,因為到最后關機重啟后,不知道是什么原因,系統的設置中,好多軟件打不開了。 建議主要看下邊Ubuntu16.04的安裝,我又重裝的,效果很好。
總體思路:
1、先裝ubuntu14.04。用UltralSO搞個刻錄U盤就好(不知道怎么回事,電腦開不了機,嘗試過16.04版本,但是感覺沒有14.04好搞。。。)
2、禁用Ubuntu自帶顯卡驅動
Ubuntu的nouveau禁用方法: 在/etc/modprobe.d中創建文件blacklist-nouveau.conf,在文件中輸入一下內容 blacklist nouveau options nouveau modeset=0
3、安裝cuda
說明:
(a)可以先安裝NVIDIA顯卡驅動,再安裝cuda,但是先裝顯卡驅動之后,就要注意,在安裝cuda時,就不要重裝cuda里帶的顯卡驅動了。
(b)我的流程就是,不自己去裝顯卡驅動,用cuda里的。
(c)在安裝cuda時,需要關閉圖形化界面
使用alt ctrl f1-f6中的任意一個,進入黑屏命令行中,使用用戶名和密碼登錄 service lightdm stop (使用root賬戶執行該命令) 將下載好的進行安裝,我這里用的是 cuda_8.0.61_375.26_linux.run 所以執行 sh cuda_8.0.61_375.26_linux.run 然后根據提示進行安裝,當遇到提示是否安裝openGL時,選擇no(不明所以,只是其他人這么說) 重啟電腦 配置環境變量 終端中輸入 $ sudo gedit /etc/profile 在打開的文件末尾,添加以下兩行。 export PATH=/usr/local/cuda-8.0/bin:$PATH export LD_LIBRARY_PATH=/usr/local/cuda-8.0/lib64:$LD_LIBRARY_PATH 使用命令 nvidia-smi 查看當前顯卡狀態
4、安裝cudnn
將cuDNN文件copy到和cuda同一目錄下,然后進行解壓,解壓之后,執行以下命令 sudo cp cuda/include/cudnn.h /usr/local/cuda/include/ sudo cp cuda/lib64/libcudnn* /usr/local/cuda/lib64/ sudo chmod a+r /usr/local/cuda/include/cudnn.h sudo chmod a+r /usr/local/cuda/lib64/libcudnn* 然后通過執行cuda中samples的deviceQuery來驗證。
5、安裝openCV 使用github上的 OpenCV安裝腳本,操作超級簡單,超級好用。(不過,可能需要很長時間,我單顆 i7 跑了將近1h)
https://github.com/jayrambhia/Install-OpenCV 最后出現OpenCV-3.3.0 ready to be used類似的話,就說明安裝成功了
6、安裝caffe
安裝依賴 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 sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev BLAS安裝 sudo apt-get install libatlas-base-dev 安裝anaconda之后pycaffe依賴好像就不用裝了,我忘了,這里附上(所有python庫依賴,本來是一行,這里為了排版,分成多行了) 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 然后從github上把caffe clone下來 git clone https://github.com/BVLC/caffe.git 然后復制一份Makefile.config文件,自己去修改(這里修改,可以參照網上的一些教程,按需修改,下邊附上我修改的) cp Makefile.config.example Makefile.config 然后在 caffe文件根目錄下進行編譯 make all -j8 (這里的 -j8 意思是使用8核進行編譯, 如果電腦是4核,用 -j4.類似,電腦有幾核,最多就可以用幾核編譯) make test -j8 make runtest -j8 進行這三步應該會遇到一些問題,請自行Google 如果出錯,要重新編譯,使用 make clean (但注意,一使用該命令,所以編譯操作就要全部重做一次) 需要使用python caffe接口,使用 make pycaffe -j8 驗證 在caffe文件根目錄下 cd python 切換到 ./python 目錄中然后在終端下輸入 python (進入python 命令行中) import caffe (如果不報錯,就說明沒問題了) 需要使用matlab caffe接口,使用 make matcaffe -j8 到這里所有編譯全部完成。
7、安裝anaconda(推薦使用anaconda,因為這里集合了python中常用的科學工具包,不用自己之后一個一個pip install,不足之處就是,可能動態鏈接庫會報錯,不過好解決。caffe的issue中有相關回答,或者直接Google)
推薦使用pycharm 配合使用,感覺超好。
pycharm中import caffe/caffe2 ---> http://blog.csdn.net/u013010889/article/details/70808866
至此,caffe 和 python環境應該就沒問題了,但具體其他操作呢??
這里推薦一個大牛的博客,自己去翻他的文章學習吧,文章的可看度還是挺高的。
http://www.cnblogs.com/denny402/tag/caffe/
matlab2017a安裝教程
http://blog.csdn.net/m0_37407756/article/details/73187654
附的Makefile.config 僅供參考,不建議直接復制
## 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 # uncomment to disable IO dependencies and corresponding data layers # USE_OPENCV := 0 # USE_LEVELDB := 0 # USE_LMDB := 0 # uncomment to allow MDB_NOLOCK when reading LMDB files (only if necessary) # You should not set this flag if you will be reading LMDBs with any # possibility of simultaneous read and write # ALLOW_LMDB_NOLOCK := 1 # Uncomment if you're using OpenCV 3 OPENCV_VERSION := 3 # 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 through *_61 lines for compatibility. # For CUDA < 8.0, comment the *_60 and *_61 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_52,code=sm_52 \ -gencode arch=compute_60,code=sm_60 \ -gencode arch=compute_61,code=sm_61 \ -gencode arch=compute_61,code=compute_61 # BLAS choice: # atlas for ATLAS (default) # mkl for MKL # open for OpenBlas BLAS := atlas # 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 # Homebrew puts openblas in a directory that is not on the standard search path # BLAS_INCLUDE := $(shell brew --prefix openblas)/include # BLAS_LIB := $(shell brew --prefix openblas)/lib # 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_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. # 注意,如果用anaconda的話,下邊兩行都要注釋掉 # PYTHON_INCLUDE := /usr/include/python2.7 \ # /usr/lib/python2.7/dist-packages/numpy/core/include # Anaconda Python distribution is quite popular. Include path: # Verify anaconda location, sometimes it's in root. # 這里anaconda的目錄根據你安裝的目錄設置,如果是按默認安裝的,因為默認安裝目錄為/root目錄下,所以為 ANACONDA_HOME := $(HOME)/anaconda ANACONDA_HOME := /home/unicoe/anaconda2 # 將下邊的#號注釋都刪除掉,共3行 PYTHON_INCLUDE := $(ANACONDA_HOME)/include \ $(ANACONDA_HOME)/include/python2.7 \ $(ANACONDA_HOME)/lib/python2.7/site-packages/numpy/core/include # Uncomment to use Python 3 (default is Python 2) # PYTHON_LIBRARIES := boost_python3 python3.5m # PYTHON_INCLUDE := /usr/include/python3.5m \ # /usr/lib/python3.5/dist-packages/numpy/core/include # We need to be able to find libpythonX.X.so or .dylib. # PYTHON_LIB := /usr/lib PYTHON_LIB := $(ANACONDA_HOME)/lib # Homebrew installs numpy in a non standard path (keg only) # PYTHON_INCLUDE += $(dir $(shell python -c 'import numpy.core; print(numpy.core.__file__)'))/include # PYTHON_LIB += $(shell brew --prefix numpy)/lib # Uncomment to support layers written in Python (will link against Python libs) WITH_PYTHON_LAYER := 1 # Whatever else you find you need goes here. INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib # If Homebrew is installed at a non standard location (for example your home directory) and you use it for general dependencies # INCLUDE_DIRS += $(shell brew --prefix)/include # LIBRARY_DIRS += $(shell brew --prefix)/lib # NCCL acceleration switch (uncomment to build with NCCL) # https://github.com/NVIDIA/nccl (last tested version: v1.2.3-1+cuda8.0) # USE_NCCL := 1 # Uncomment to use `pkg-config` to specify OpenCV library paths. # (Usually not necessary -- OpenCV libraries are normally installed in one of the above $LIBRARY_DIRS.) # USE_PKG_CONFIG := 1 # N.B. both build and distribute dirs are cleared on `make clean` BUILD_DIR := build DISTRIBUTE_DIR := distribute # Uncomment for debugging. Does not work on OSX due to https://github.com/BVLC/caffe/issues/171 # DEBUG := 1 # The ID of the GPU that 'make runtest' will use to run unit tests. TEST_GPUID := 0 # enable pretty build (comment to see full commands) Q ?= @
之后的錯誤記錄,可能還有遺漏,自行Google吧
裝完anaconda2之后,記得設置一下環境變量
export PATH="/home/tom/anaconda2/bin:$PATH"
./build/tools/caffe: error while loading shared libraries: libhdf5_hl.so.8: cannot open shared object file: No such file or directory
在 /etc/profile 加入環境變量
export LD_LIBRARY_PATH="/usr/local/cuda/lib64"
export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/home/xxx/anaconda2/lib"
注意,這里的 /home/xxx/anaconda2 是你裝anaconda2的目錄
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "caffe/__init__.py", line 1, in <module>
from .pycaffe import Net, SGDSolver, NesterovSolver, AdaGradSolver, RMSPropSolver, AdaDeltaSolver, AdamSolver, NCCL, Timer
File "caffe/pycaffe.py", line 15, in <module>
import caffe.io
File "caffe/io.py", line 8, in <module>
from caffe.proto import caffe_pb2
File "caffe/proto/caffe_pb2.py", line 4, in <module>
from google.protobuf.internal import enum_type_wrapper
ImportError: No module named google.protobuf.internal
解決方法:
我是用anaconda2/bin 目錄下的pip 進行 pip install protobuf 解決問題
安裝Ubuntu16.04 caffe 和 py faster rcnn是參考一下幾篇博客搞定的
用當前最新的 Ubuntu16.04.3
Ubuntu16.04 Caffe 安裝步驟記錄(超詳盡) http://blog.csdn.net/yhaolpz/article/details/71375762
我的做法:
1、安裝完Ubuntu之后,先換顯卡驅動,
(a)換顯卡之前,先執行 sudo apt-get update sudo apt-get upgrade ,將系統更新之后, 在System Settings -> Software & Updates -> Additional Drivers 中選擇要換的顯卡驅動(我當前的是NVIDIA-384),安裝完成之后。照着上邊的教程把默認顯卡驅動禁用了,然后重啟查看有沒有問題(看能不能進入系統,能進入的話,再進行后續步驟)
之后我的步驟是
(b)安裝依賴包,
(c)配置環境變量,
(d)安裝cuda(這里要注意的是,因為不用cuda帶的驅動,所以安裝的時候,不用關閉圖形化界面),
(e)安裝cudnn也沒用他的方法,我是將cudnn安裝包放在了/usr/local/ 目錄下,直接解壓,然后用我博客上邊的操作,然后用nvcc -V驗證一下。
(f)裝openCV也是直接用安裝腳本,直接搞定,自己做的很少。 (裝完之后,重啟看看)
(g)裝caffe(我裝在了 /home/unicoe/ 目錄下,所以要用 sudo make all -j8 && sudo make runtest -j8 )
(h)裝anaconda2(注意,在裝caffe的時候,不要用anaconda2,因為會有蜜汁錯誤,裝了caffe之后再裝也是一樣的)
(i)裝pycaffe(裝了anaconda2之后,配置一下環境變量,就不用想博主那樣裝一堆Python庫了)
(j)進入 caffe/python 中 python 然后 import caffe,這里會出現一些問題,總結來說,就是缺啥裝啥
好像要裝 conda install easydict
conda install opencv-python
Traceback (most recent call last)
File ImportError: /home/../anaconda2/lib/python2.7/site-packages/zmq/backend/cython/../../../../.././libstdc++.so.6: versionGLIBCXX_3.4.21' not found
解決:
https://github.com/BVLC/caffe/issues/4953
https://gitter.im/BVLC/caffe/archives/2015/08/20
cd /home/unicoe/caffe
pip install protobuf
sudo apt-get install python-protobuf
conda install libgcc
如果還有其他錯誤,自行Google吧
裝faster rcnn
參照 http://blog.csdn.net/u012841667/article/details/53436615#reply
(a)先下載 py-faster-rcnn
git clone --recursive https://github.com/rbgirshick/py-faster-rcnn.git
(b) cd py-faster-rcnn/lib 然后 make
- cd py-faster-rcnn/lib
- make
(c)編譯/py-faster-rcnn/caffe-fast-rcnn
改動的地方,和裝caffe時一樣
(I)改Makefile.config (這個可以復制裝完的caffe中的Makefile.config)
(II)改Makefile(這個手動改,照着裝caffe時的方法,不要直接復制caffe中的Makefile文件,不然會有各種問題)
(d) make -j8 && make pycaffe
編譯的時候,就會報關於cudnn的錯,上邊的博客中說了,是這里的cudnn版本低,需要換成,現在caffe版本的cudnn,
(e)參照上邊的博客,cudnn依賴要改動三個地方(使用 cp命令 cp 源文件位置/源文件 目的文件位置, 具體使用,請自行查看)
(I)將/py-faster-rcnn/caffe-fast-rcnn/include/caffe/util/cudnn.hpp 換成最新版的caffe里的cudnn的實現,即相應的cudnn.hpp
(II)將/py-faster-rcnn/caffe-fast-rcnn/src/caffe/layer里的,所有以cudnn開頭的文件,例如cudnn_lrn_layer.cu,cudnn_pooling_layer.cpp,cudnn_sigmoid_layer.cu。 都替換成最新版的caffe里的相應的同名文件
(III)將./include/caffe/layers的,所有以cudnn開頭的文件,例如cudnn_conv_layer.hpp,cudnn_lcn_laye.hpp 都替換成最新版的caffe里的相應的同名文件
(f)然后就可以運行demo.py 了
- cd py-faster-rcnn/tools
- ./demo.py
如果有下列錯誤,請參照下邊博客中所說
OSError: libcudnn.so.7.0: cannot open shared object file: No such file or directory錯誤
http://blog.csdn.net/u014696921/article/details/60140264
可供參考的blog還有 Ubuntu16.04 caffe安裝記錄 http://www.cnblogs.com/peiyuYang/p/7784787.html
如有問題,請留言,或者發郵件到unicoe@163.com 中。研究生之路剛開始,希望能和大家多多交流。
