In [24]:

    

from numba import jit


@jit
def f(x, y):
    return x + y


    

In [25]:

    

%timeit f(1, 2)


    

    




The slowest run took 132984.30 times longer than the fastest. This could mean that an intermediate result is being cached.
1000000 loops, best of 3: 202 ns per loop

In [8]:

    

f(1j, 2)


    

    
Out[8]:





(2+1j)

In [20]:

    

from numba import int32

@jit(int32(int32, int32))
def f1(x, y):
    return x + y


    

In [22]:

    

%timeit f1(1,2)


    

    




The slowest run took 29.61 times longer than the fastest. This could mean that an intermediate result is being cached.
1000000 loops, best of 3: 213 ns per loop

In [14]:

    

f1(1j, 2)


    

    




---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-14-19300b83c20f> in <module>()
----> 1 f1(1j, 2)

C:\ProgramData\Anaconda3\lib\site-packages\numba\dispatcher.py in _explain_matching_error(self, *args, **kws)
    397         msg = ("No matching definition for argument type(s) %s"
    398                % ', '.join(map(str, args)))
--> 399         raise TypeError(msg)
    400 
    401     def _search_new_conversions(self, *args, **kws):

TypeError: No matching definition for argument type(s) complex128, int64

In [23]:

    

%timeit f(2**32, 2**31 + 1)


    

    




The slowest run took 12.92 times longer than the fastest. This could mean that an intermediate result is being cached.
1000000 loops, best of 3: 275 ns per loop

In [26]:

    

!ping www.bbc.co.uk


    

    




正在 Ping www.bbc.net.uk [212.58.246.93] 具有 32 字节的数据:
来自 212.58.246.93 的回复: 字节=32 时间=280ms TTL=46
来自 212.58.246.93 的回复: 字节=32 时间=318ms TTL=46
来自 212.58.246.93 的回复: 字节=32 时间=311ms TTL=46
来自 212.58.246.93 的回复: 字节=32 时间=300ms TTL=46

212.58.246.93 的 Ping 统计信息:
    数据包: 已发送 = 4，已接收 = 4，丢失 = 0 (0% 丢失)，
往返行程的估计时间(以毫秒为单位):
    最短 = 280ms，最长 = 318ms，平均 = 302ms

In [27]:

    

from IPython.display import display


    

In [28]:

    

display(print)


    

    






<function print>

In [31]:

    

import numpy as np
from numba import generated_jit, types

@ generated_jit(nopython=True)
def is_missing(x):
    if isinstance(x, types.Float):
        return lambda x: np.isnan(x)
    elif isinstance(x, types.NPDatetime, types.NPTimedelta):
        missing = x(NaT)
        return lambda x: x == missing
    else:
        return lambda: False


    

In [34]:

    

is_missing(np.NaN)


    

    
Out[34]:





True

In [36]:

    

from numba import vectorize, float64

@vectorize([float64(float64, float64)])
def f3(x, y):
    return x + y


    

In [38]:

    

f3(2.0, 3.33)


    

    
Out[38]:





5.3300000000000001

In [42]:

    

from numba import float64, int64, float32

@vectorize([int32(int32, int32),
            int64(int64, int64),
            float32(float32, float32),
            float64(float64, float64)])
def f(x, y):
    return x + y


    

In [48]:

    

a = np.arange(6,dtype='float64')
a


    

    
Out[48]:





array([ 0.,  1.,  2.,  3.,  4.,  5.])

In [49]:

    

f(a,a)


    

    
Out[49]:





array([  0.,   2.,   4.,   6.,   8.,  10.])

In [50]:

    

a = np.linspace(0, 1+1j, 6)
a


    

    
Out[50]:





array([ 0.0+0.j ,  0.2+0.2j,  0.4+0.4j,  0.6+0.6j,  0.8+0.8j,  1.0+1.j ])

In [51]:

    

f(a,a)


    

    




---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-51-12d30bce134a> in <module>()
----> 1 f(a,a)

TypeError: ufunc 'f' not supported for the input types, and the inputs could not be safely coerced to any supported types according to the casting rule ''safe''

In [52]:

    

a = np.arange(12).reshape(3,4)


    

In [53]:

    

a


    

    
Out[53]:





array([[ 0,  1,  2,  3],
       [ 4,  5,  6,  7],
       [ 8,  9, 10, 11]])

In [54]:

    

f(a, a)


    

    
Out[54]:





array([[ 0,  2,  4,  6],
       [ 8, 10, 12, 14],
       [16, 18, 20, 22]])

In [55]:

    

f.reduce(a, axis=0)  # 降维


    

    
Out[55]:





array([12, 15, 18, 21])

In [56]:

    

f.reduce(a, axis=1)


    

    
Out[56]:





array([ 6, 22, 38])

In [58]:

    

f.accumulate(a, axis=1)


    

    
Out[58]:





array([[ 0,  1,  3,  6],
       [ 4,  9, 15, 22],
       [ 8, 17, 27, 38]])

In [59]:

    

f.accumulate(a)


    

    
Out[59]:





array([[ 0,  1,  2,  3],
       [ 4,  6,  8, 10],
       [12, 15, 18, 21]])

In [63]:

    

from numba import guvectorize

@guvectorize([(int64[:], int64[:], int64[:])], '(n),()->(n)', nopython=True)
def g(x, y, res):
    for i in range(x.shape[0]):
        res[i] = x[i] + y[0]


    

In [65]:

    

a = np.arange(6).reshape(2, 3)
g(a, 10)


    

    
Out[65]:





array([[10, 11, 12],
       [13, 14, 15]], dtype=int64)

In [66]:

    

from numba import cfunc

@cfunc("float64(float64, float64)")
def add(x, y):
    return x + y


    

In [67]:

    

@cfunc("float64(float64, float64)")
def add(x, y):
    return x + y

print(add.ctypes(4.0, 5.0))  # prints "9.0"


    

    




9.0

In [68]:

    

from numba.pycc import CC

cc = CC('my_module')
# Uncomment the following line to print out the compilation steps
#cc.verbose = True

@cc.export('multf', 'f8(f8, f8)')
@cc.export('multi', 'i4(i4, i4)')
def mult(a, b):
    return a * b

@cc.export('square', 'f8(f8)')
def square(a):
    return a ** 2
cc.compile()


    

In [69]:

    

import my_module
my_module.multi(3, 4)


    

    
Out[69]:





12

In [70]:

    

my_module.square(1.414)


    

    
Out[70]:





1.9993959999999997

In [72]:

    

f.inspect_types()


    

    




---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-72-d5cd027f6f78> in <module>()
----> 1 f.inspect_types()

AttributeError: 'DUFunc' object has no attribute 'inspect_types'

In [ ]: