在GCP安裝python3，運作pytorch

1. 1.安裝python3
2. 2.安裝cuda
3. 3.安裝cudnn
4. 4.安裝tensorflow
5. 5.安裝常用套件
6. 6.小記錄

好吧，這篇說是十二月中寫的，但實際上二月初我才發布…老實說我快忘了要保持一個月至少一篇日誌的習慣，雖然說墮落是人類的原罪，但為什麼人非得汲汲營營呢？嗯，因為這是個無解的命題，所以我就發發牢騷，以下會記錄我當初如何在GCP開啟pytorch並實現python3的環境，當初找資料找了許久，因為GCP的ubuntu預設是python2，對於伸手黨來說大部分的資料都是python3…。

1.安裝python3

sudo add-apt-repository ppa:fkrull/deadsnakes
sudo apt-get update
sudo apt-get install python3.5
sudo apt-get install python3.5-dev
sudo apt-get install libncurses5-dev
wget https://bootstrap.pypa.io/get-pip.py
sudo python3.5 get-pip.py
# 修改/usr/bin/ 底下的路徑
sudo ln -s /usr/bin/python3.5 /usr/local/bin/python3
sudo ln -s /usr/local/bin/pip /usr/local/bin/pip3

2.安裝cuda

cuda(https://cloud.google.com/compute/docs/gpus/add-gpus)

#!/bin/bash
echo "Checking for CUDA and installing."
# Check for CUDA and try to install.
if ! dpkg-query -W cuda-8-0; then
  curl -O http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1404/x86_64/cuda-repo-ubuntu1404_8.0.61-1_amd64.deb
  dpkg -i ./cuda-repo-ubuntu1404_8.0.61-1_amd64.deb
  apt-get update
  apt-get install cuda-8-0 -y
  apt-get install linux-headers-$(uname -r) -y
fi
# Enable persistence mode

nvidia-smi -pm 1

修改路徑

# ~/.bashrc 加入
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda/lib64:/usr/local/cuda/extras/CUPTI/lib64:/usr/lib/nvidia-367
export CUDA_HOME=/usr/local/cuda
export PATH=$PATH:/usr/local/cuda/bin

source ~/.bashrc

3.安裝cudnn

先下載cudnn-8.0-linux-x64-v5.1.tgz

tar -zxvf cudnn-8.0-linux-x64-v5.1.tgz
cd cuda
sudo cp include/* /usr/local/cuda-8.0/include/
sudo cp lib64/* /usr/local/cuda-8.0/lib64/

4.安裝tensorflow

這裡要注意版本問題，至於為何要注意…我記得好像是因為執行會有問題？

sudo pip install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.2.1-cp35-cp35m-linux_x86_64.whl
sudo pip install https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.2.1-cp35-cp35m-linux_x86_64.whl

5.安裝常用套件

sudo pip install tqdm
sudo pip install numpy
sudo pip install pandas
sudo pip install scikit-learn
sudo pip install scipy
sudo pip install pymongo
sudo pip install http://download.pytorch.org/whl/cu80/torch-0.3.0.post4-cp35-cp35m-linux_x86_64.whl
sudo pip install torchvision
sudo pip install dill
sudo pip install opencv-python
sudo pip install cupy
sudo pip install pyinn 
...

6.小記錄

啊，看到當初截的圖，才想起來當時一口氣開了五顆GPU，一晚就能燒掉一百多美金，要不是有各項比賽的免費兌換券，根本無法支持這樣燒錢的玩法呢~

To be continued..