In [1]:

    

import skcuda.fft as cu_fft


    

In [2]:

    

import skcuda.cusolver as solver


    

In [ ]:

    




    

In [3]:

    

# %load https://raw.github.com/lebedov/scikits.cuda/master/demos/fft_demo.py
#!/usr/bin/env python

"""
Demonstrates how to use the PyCUDA interface to CUFFT to compute 1D FFTs.
"""

import pycuda.autoinit
import pycuda.gpuarray as gpuarray
import numpy as np

import skcuda.fft as cu_fft

print 'Testing fft/ifft..'
N = 4096*16

x = np.asarray(np.random.rand(N), np.float32)
xf = np.fft.fft(x)
y = np.real(np.fft.ifft(xf))

x_gpu = gpuarray.to_gpu(x)
xf_gpu = gpuarray.empty(N/2+1, np.complex64)
plan_forward = cu_fft.Plan(x_gpu.shape, np.float32, np.complex64)
cu_fft.fft(x_gpu, xf_gpu, plan_forward)

y_gpu = gpuarray.empty_like(x_gpu)
plan_inverse = cu_fft.Plan(x_gpu.shape, np.complex64, np.float32)
cu_fft.ifft(xf_gpu, y_gpu, plan_inverse, True)

print 'Success status: ', np.allclose(y, y_gpu.get(), atol=1e-6)

print 'Testing in-place fft..'
x = np.asarray(np.random.rand(N)+1j*np.random.rand(N), np.complex64)
x_gpu = gpuarray.to_gpu(x)

plan = cu_fft.Plan(x_gpu.shape, np.complex64, np.complex64)
cu_fft.fft(x_gpu, x_gpu, plan)

cu_fft.ifft(x_gpu, x_gpu, plan, True)

print 'Success status: ', np.allclose(x, x_gpu.get(), atol=1e-6)


    

    




Testing fft/ifft..
Success status:  True
Testing in-place fft..
Success status:  True

In [4]:

    

import skcuda


    

In [5]:

    

skcuda.__version__


    

    
Out[5]:





'0.5.1'

In [6]:

    

skcuda.cusolver.cusolver


    

    





AttributeErrorTraceback (most recent call last)
<ipython-input-6-fcfe6107e98b> in <module>()
----> 1 skcuda.cusolver.cusolver

AttributeError: 'module' object has no attribute 'cusolver'

In [ ]:

    




    

In [7]:

    

%%time 
cu_fft.fft(x_gpu, x_gpu, plan)
x_gpu.get()[1234]


    

    




CPU times: user 0 ns, sys: 4 ms, total: 4 ms
Wall time: 666 µs

In [8]:

    

%%time 
xfft = np.fft.fft(x)


    

    




CPU times: user 8 ms, sys: 0 ns, total: 8 ms
Wall time: 5.83 ms

In [9]:

    

x.shape[0]/1024**2


    

    
Out[9]:





0

In [10]:

    

x_gpu.get()[1234]


    

    
Out[10]:





(32.306984+105.99432j)

In [11]:

    

# %load https://github.com/lebedov/scikits.cuda/raw/master/demos/dot_demo.py
#!/usr/bin/env python

"""
Demonstrates multiplication of two matrices on the GPU.
"""

import pycuda.autoinit
import pycuda.gpuarray as gpuarray
import pycuda.driver as drv
import numpy as np

import skcuda.linalg as culinalg
import skcuda.misc as cumisc
culinalg.init()

# Double precision is only supported by devices with compute
# capability >= 1.3:
import string
demo_types = [np.float32, np.complex64]
if cumisc.get_compute_capability(pycuda.autoinit.device) >= 1.3:
    demo_types.extend([np.float64, np.complex128])

for t in demo_types:
    print 'Testing matrix multiplication for type ' + str(np.dtype(t))
    if np.iscomplexobj(t()):
        a = np.asarray(np.random.rand(10, 5)+1j*np.random.rand(10, 5), t)
        b = np.asarray(np.random.rand(5, 5)+1j*np.random.rand(5, 5), t)
        c = np.asarray(np.random.rand(5, 5)+1j*np.random.rand(5, 5), t)
    else:
        a = np.asarray(np.random.rand(10, 5), t)
        b = np.asarray(np.random.rand(5, 5), t)
        c = np.asarray(np.random.rand(5, 5), t)

    a_gpu = gpuarray.to_gpu(a)
    b_gpu = gpuarray.to_gpu(b)
    c_gpu = gpuarray.to_gpu(c)

    temp_gpu = culinalg.dot(a_gpu, b_gpu)
    d_gpu = culinalg.dot(temp_gpu, c_gpu)
    temp_gpu.gpudata.free()
    del(temp_gpu)
    print 'Success status: ', np.allclose(np.dot(np.dot(a, b), c) , d_gpu.get())

    print 'Testing vector multiplication for type '  + str(np.dtype(t))
    if np.iscomplexobj(t()):
        d = np.asarray(np.random.rand(5)+1j*np.random.rand(5), t)
        e = np.asarray(np.random.rand(5)+1j*np.random.rand(5), t)
    else:
        d = np.asarray(np.random.rand(5), t)
        e = np.asarray(np.random.rand(5), t)

    d_gpu = gpuarray.to_gpu(d)
    e_gpu = gpuarray.to_gpu(e)

    temp = culinalg.dot(d_gpu, e_gpu)
    print 'Success status: ', np.allclose(np.dot(d, e), temp)


    

    




Testing matrix multiplication for type float32
Success status:  True
Testing vector multiplication for type float32
Success status:  True
Testing matrix multiplication for type complex64
Success status:  True
Testing vector multiplication for type complex64
Success status:  True
Testing matrix multiplication for type float64
Success status:  True
Testing vector multiplication for type float64
Success status:  True
Testing matrix multiplication for type complex128
Success status:  True
Testing vector multiplication for type complex128
Success status:  True

In [ ]:

    




    

In [47]:

    




    

In [ ]: