開源機器人俱樂部線上聚會 - 2018年6月7日星期四中午12點 (Ubuntu 16.04, CUDA, cuDNN, Intel Python, TensorFlow 安裝教學,以 GTX-1080Ti 為例)

大家好!俱樂部聚會這次以「臉書線上直播」方式進行,內容如下:

 

1. 最新機器人新聞及技術分享

 

2. Ubuntu 16.04, CUDA, cuDNN, Intel Python, TensorFlow 安裝教學,以 GTX-1080Ti 為例

 

3. 會員線上文字交流

 

主持人:李明達老師

 

 

 

 

[Upgrade BIOS]

 

[System settings-> Brightness & Lock -> Never, OFF, Uncheck]

 

[Lock Terminal to luncher]

 

[Connect Ethernet or Wi-Fi]

 

sudo apt-get update

sudo apt-get dist-upgrade

 

[reboot]

 

 

https://medium.com/@ishengfang/%E5%A6%82%E4%BD%95%E5%9C%A8ubuntu-17-04%E6%9E%B6%E8%A8%ADdeep-learning-%E7%92%B0%E5%A2%83-ca7377d98d5d

 

sudo apt install htop

sudo apt install gcc

sudo apt install make

sudo apt install git

 

sudo add-apt-repository ppa:graphics-drivers/ppa

sudo apt-get update

sudo apt-get install nvidia-410

 

[reboot]

 

 

cd OpenDBT_Linux64/tool/NVIDIA/

 

sudo dpkg -i cuda-repo-ubuntu1604-9-0-local_9.0.176-1_amd64.deb

sudo apt-key add /var/cuda-repo-9-0-local/7fa2af80.pub

sudo apt-get update

sudo apt-get install cuda

 

 

sudo dpkg -i cuda-repo-ubuntu1604-9-0-local-cublas-performance-update_1.0-1_amd64.deb

sudo apt-get update

 

sudo dpkg -i cuda-repo-ubuntu1604-9-0-local-cublas-performance-update-2_1.0-1_amd64.deb

sudo apt-get update

 

sudo dpkg -i cuda-repo-ubuntu1604-9-0-local-cublas-performance-update-3_1.0-1_amd64.deb

sudo apt-get update

 

sudo dpkg -i libcudnn7_7.0.5.15-1+cuda9.0_amd64.deb

sudo apt-get update

sudo dpkg -i libcudnn7-dev_7.0.5.15-1+cuda9.0_amd64.deb sudo apt-get update

 

 

sudo dpkg -i libcudnn7-doc_7.0.5.15-1+cuda9.0_amd64.deb

sudo apt-get update

 

sudo apt-get install cuda-command-line-tools-9-0 sudo apt-get update

 

 

 

[Download Anaconda for Linux]

 

bash Anaconda3-5.2.0-Linux-x86_64.sh

yes

yes

yes

 

[Reboot]

 

conda update conda

 

 

[Installing the Intel Distribution for Python]

 

https://software.intel.com/en-us/articles/using-intel-distribution-for-python-with-anaconda

 

conda config --add channels intel

 

conda create -n idp intelpython3_full python=3.5

 

source activate idp

 

pip install --upgrade pip

 

[Intel]

 

conda install ipp

 

conda install mkl

 

conda install tbb

 

conda install daal

 

conda install numpi

 

[Installing TensorFlow]

 

https://www.tensorflow.org/install/install_linux

 

cd

vi .bashrc

export PATH=/usr/local/cuda-9.0/bin${PATH:+:${PATH}} export LD_LIBRARY_PATH=/usr/local/cuda-9.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}} export LD_LIBRARY_PATH=${LD_LIBRARY_PATH:+${LD_LIBRARY_PATH}:}/usr/local/cuda/extras/CUPTI/lib64

 

 

[GPU]

pip install --ignore-installed --upgrade

https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.8.0-cp35-cp35m-linux_x86_64.whl

 

 

[CPU]

pip install --ignore-installed --upgrade https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.8.0-cp35-cp35m-linux_x86_64.whl

 

 

 

[Install keras]

source activate idp

pip install keras==2.1.5  (If 'Permission denied', type sudo -s)

 

[Install tqdm]

source activate idp

pip install tqdm

 

[Install OpenCV]

source activate idp

pip install opencv-python

 


延伸閱讀

1.
OpenDBT - 開源大補帖

2.
OpenR8 - 人人都會用的 AI 軟體 (下載)

3.
開源機器人俱樂部線上聚會 - 2018年6月1日星期五中午12點

4.
[ezAI 簡單學AI] PCB 電路板的深度學習影像分析

[CP 值最好的 AI 人工智慧電腦熱賣中]
硬體採用最新最快的 Titan RTX, RTX-2080Ti-11G, 及 TESLA V100!
視窗圖控介面,易學易用。
硬體優勢:針對深度學習最佳化設計,且經過客戶端 24H 運作驗證。
軟體優勢:事先安裝人工智慧軟體及優化設定,每週提供最新 AI 演算法下載。
公司優勢:專業人工智慧機器人設計公司,資本額一億。
歡迎企業、學校、及公家機關大量採購。
可租用,歡迎來信或來電洽詢。
詳細資料請參閱以下網址: https://tw.leaderg.com/ai_computer