Notebook

Adding two Scalars¶

In [1]:

import theano.tensor as T

In [3]:

from theano import function

In [4]:

x = T.dscalar('x')

In [5]:

y = T.dscalar('y')

In [6]:

z = x + y

In [7]:

f = function([x, y], z)

==========================

In [8]:

f(2, 3)

Out[8]:

array(5.0)

In [9]:

f(16.3, 12.1)

Out[9]:

array(28.4)

T.dscalar is the type we assign to “0-dimensional arrays (scalar) of doubles (d)” It is a Theano Type. dscalar is not a class. Therefore, neither x nor y are actually instances of dscalar. They are instances of TensorVariable. x and y are, however, assigned the theano Type dscalar in their type field

In [10]:

type(x)

Out[10]:

theano.tensor.var.TensorVariable

In [11]:

x.type

Out[11]:

TensorType(float64, scalar)

In [12]:

T.dscalar

Out[12]:

TensorType(float64, scalar)

In [16]:

x.type is T.dscalar

Out[16]:

True

You can use the pp function to pretty-print out the computation associated to z.

In [17]:

from theano import pp
print pp(z)

(x + y)

In [18]:

f = function([x, y], z)

The first argument([x,y]) is inputs. The second argment(z) is output

In [19]:

z.eval({x:16.3, y:12.1})

Out[19]:

array(28.4)

Adding tow Matrices¶

In [20]:

x = T.dmatrix('x')
y = T.dmatrix('y')
z = x + y
f = function([x, y], z)

In [21]:

f([[1, 2], [3, 4]], [[10, 20], [30, 40]])

Out[21]:

array([[ 11.,  22.],
       [ 33.,  44.]])

In [22]:

import numpy
f(numpy.array([[1, 2], [3, 4]]), numpy.array([[10, 20], [30, 40]]))

Out[22]:

array([[ 11.,  22.],
       [ 33.,  44.]])

byte : bscalar, bvector, bmatrix, brow, bcol, btensor3, btensor4 16-bit integers: wscalar, wvector, wmatrix, wrow, wcol, wtensor3, wtensor4 32-bit integers: iscalar, ivector, imatrix, irow, icol, itensor3, itensor4 64-bit integers: lscalar, lvector, lmatrix, lrow, lcol, ltensor3, ltensor4 float : fscalar, fvector, fmatrix, frow, fcol, ftensor3, ftensor4 double : dscalar, dvector, dmatrix, drow, dcol, dtensor3, dtensor4 complex : cscalar, cvector, cmatrix, crow, ccol, ctensor3, ctensor4

In [37]:

i = T.iscalar('i')
l = T.lscalar('l')
j = T.scalar('j')

In [28]:

i.type

Out[28]:

TensorType(int32, scalar)

In [29]:

l.type

Out[29]:

TensorType(int64, scalar)

In [38]:

j.type

Out[38]:

TensorType(float64, scalar)

In [30]:

a = T.vector()
out = a + a ** 10
f = function([a],out)
print f([0,1,2])

[    0.     2.  1026.]

In [36]:

b = T.vector()
out2 = a**2 + b**2 + 2*a*b
f2 = function([a,b], out2)
print f2([1,2,3],[2,3,4])

[  9.  25.  49.]

In [ ]: