1. 程式人生 > >深度學習筆記整理-pycharm環境下開發及執行

深度學習筆記整理-pycharm環境下開發及執行

1. 關於軟體安裝: Anaconda---->>>安裝python3.6 package---->>>安裝pycharm IDE------->>>配置interpretation

                                              ----->>>安裝tensorflow backend------>>>安裝theano------->>>安裝keras、pandas等(根據需求tensorflow和theano二選一)

2. 新建project:配置interpretation------>>>測試程式碼 執行 

3. 遇到的問題:

     1) theano提示g++.exe檔案無法發現,由於tensorflow需要c編譯器,呼叫CPU和GPU

WARNING (theano.configdefaults): g++ not detected ! Theano will be unable to execute optimized C-implementations (for both CPU and GPU) and will default to Python implementations. Performance will be severely degraded. To remove this warning, set Theano flags cxx to an empty string.
    解決方法:
conda install mingw libpython
     2)tensorflow會優先使用全部GPU,給系統預留200M

“Your CPU supports instructions that this TensorFlow binary was not compiled”

    解決辦法:

設定提示資訊的等級:在import tensorflow 前加上   

import os
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'   #只顯示warning和errors