Windows上带有Anaconda的Theano:如何设置BLAS? [英] Theano with Anaconda on Windows: how to setup BLAS?
问题描述
我已经使用Anaconda在Windows 7 64bit上安装Theano(和Keras). 这是我的步骤.
I've used Anaconda to install Theano (and Keras) on Windows 7 64bit. Here are my steps.
- 为Python 3.5安装最新的Anaconda
- conda安装mingw libpython
- pip安装Theano
- conda安装pydot-ng
- pip install keras
- 编辑.keras/keras.json以使用"theano"代替"tensorflow".
打开Jupyter，复制并粘贴以下代码: https://github.com/fchollet/keras/blob/master/examples/cifar10_cnn.py 它可以正常执行，直到对model.fit的调用:导入，数据下载，模型编译全部工作为止.
Open Jupyter, copy and paste this code: https://github.com/fchollet/keras/blob/master/examples/cifar10_cnn.py It executes fine until the call to model.fit : imports, data download, model compilation all work.
请注意，我将data_augmentation设置为False.
Please note that I set data_augmentation to False.
对model.fit的调用失败，并显示以下错误消息: http://pastebin.com/Rq7HTc4H
The call to model.fit fails with the following error message:http://pastebin.com/Rq7HTc4H
这似乎是问题所在:
AssertionError:AbstractConv2d Theano优化失败:没有 可用的实现支持所请求的选项.你是否 从优化程序中同时排除"conv_dnn"和"conv_gemm"?如果在GPU上， cuDNN是否可用，GPU是否支持它?如果在CPU上，您有 安装了Theano的BLAS库是否可以链接?
AssertionError: AbstractConv2d Theano optimization failed: there is no implementation available supporting the requested options. Did you exclude both "conv_dnn" and "conv_gemm" from the optimizer? If on GPU, is cuDNN available and does the GPU support it? If on CPU, do you have a BLAS library installed Theano can link against?
我自己没有安装BLAS，我以为那是Anaconda的一部分. 浏览互联网会发现类似的问题. 显然，您可以通过编辑.theanorc.txt文件并添加"optimizer = None"来阻止Theano寻找BLAS库:
I didn't install BLAS myself, I thought that would be a part of Anaconda. Browsing the internet reveals similar problems. Apparently you can prevent Theano from looking for a BLAS library by editing a .theanorc.txt file and adding "optimizer = None": How configure theano on Windows?
我执行了os.path.expanduser('~/.theanorc.txt')，验证了我的.theanorc.txt在正确的位置，添加了optimier = None行并再次尝试了该代码.
没有效果.
I executed os.path.expanduser('~/.theanorc.txt'), verified that my .theanorc.txt is in the right place, added the optimier=None line and tried the code again.
No effect.
如何配置Theano在不使用BLAS的情况下工作? 或者，甚至更好的是，如何安装BLAS库(最好通过Anaconda)?
How can I configure Theano to work without BLAS ? Or, even better, how can I install a BLAS library (preferably via Anaconda) ?
更新: 显然，您可以在运行时编辑配置:
UPDATE: Apparently you can edit the config during runtime:
import theano
theano.config.optimizer="None"
消除了问题.
这意味着Theano找不到我的配置文件. 我以为我把它放到了正确的位置，是什么导致Theano找不到它?
This means that Theano doesn't find my config file. I thought I had put it at the correct location, what might prevent Theano from finding it ?
推荐答案
应为os.path.expanduser('~/.theanorc')，不带txt.
使用环境变量更改theano参数:
从您的代码开始，使用它来配置theano:
use enviroment variable to change theano parameters:
beginning on your code use this to config theano:
import os
os.environ['THEANO_FLAGS'] = 'optimizer=None'
您也可以尝试以下操作: http://deeplearning.net/software/theano/troubleshooting.html#test- blas
also you can try this: http://deeplearning.net/software/theano/troubleshooting.html#test-blas
这篇关于Windows上带有Anaconda的Theano:如何设置BLAS?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！
