In [34]:

    

import theano
from theano import tensor as T
import numpy as np
%matplotlib inline
import matplotlib.pyplot as plt
import pandas as pd


    

In [14]:

    

trX = np.linspace(-1, 1, 101)
trY = 2 * trX + np.random.randn(*trX.shape) * 0.33


    

In [15]:

    

X = T.scalar()
Y = T.scalar()


    

In [16]:

    

def model(X, w):
    return X * w


    

In [17]:

    

w = theano.shared(np.asarray(0., dtype=theano.config.floatX))
y = model(X, w)


    

In [18]:

    

print w.get_value()


    

    




0.0

In [19]:

    

cost = T.mean(T.sqr(y - Y))
gradient = T.grad(cost=cost, wrt=w)
updates = [[w, w - gradient * 0.01]]


    

In [20]:

    

train = theano.function(inputs=[X, Y], outputs=cost, updates=updates, allow_input_downcast=True)


    

In [26]:

    

costs = []


    

In [27]:

    

for x, y in zip(trX, trY):
    costs.append(train(x, y))


    

In [35]:

    

df = pd.DataFrame(costs)


    

In [38]:

    

df.plot(title='cost')


    

    
Out[38]:





<matplotlib.axes._subplots.AxesSubplot at 0x10cef77d0>