>> import numpy # OK! >>> import scipy # OK! >>> import theano # warning WARNING (theano.tensor.blas): Failed to import ...">

Install anano "blas error"

>>> import numpy # OK! >>> import scipy # OK! >>> import theano # warning WARNING (theano.tensor.blas): Failed to import scipy.linalg.blas, and Theano flag blas.ldflags is empty. Falling back on slower implementations for dot(matrix, vector), dot(vector, matrix) and dot(vector, vector) (DLL load failed: Belirtilen modül bulunamadı.) >>> 

When importing theano, python gives this warning. I configure anano at this link. http://deeplearning.net/software/theano/install_windows.html

Import to Numpy and Scipy is fine. but am i stuck in blas.?

I cloned my OpenBLAS computer from github. How do I add blas to environment variables ("path")?

+6
source share
1 answer

Apparently you have an error from scipy.linalg.blas, try installing the package from to install scipy, please make sure you have installed

NumPy + MKL

You can find it at http://www.lfd.uci.edu/~gohlke/pythonlibs/#numpy


Another problem that you mentioned is that the g ++ compiler was not detected; to compile the generated C code, you need to use g ++ (Linux and Windows). (It’s just that Google will help you deeply study anano tuning).

The training part of the MinGW installation:

  • In the C: \ folder, the name of the folder named Program
  • Install MinGW in this folder with the program file
  • Install msys in the bin folder.
  • run sh /postinstall/pi.sh in mysy, and the direction you just set MinGW should be C:/Program/mingw-w64/mingw64 (depends on the installation direction)
  • Follow the instructions at http://rosinality.ncity.net/doku.php?id=python:installing_theano and maybe your code will work!
+1
source

All Articles