Talk Is Cheap

import theano import numpy as npones = theano.shared(np.float32([[1,2,3],[4,5,6],[7,8,9]]))print(ones.get_value())print(ones.reshape([1,-1]).eval())print(ones.reshape((1,-1)).eval())print(ones.reshape(

import theano import numpy as np ones = theano.shared(np.float32([[1,2,3],[4,5,6],[7,8,9]])) temp = ones.dimshuffle([0,1,'x']) print(temp.eval()) temp = ones.dimshuffle([0,1]) print(temp.eval()) temp =

import theano import numpy as np import theano.tensor as T ones = theano.shared(np.float32([[1,2,3],[4,5,6],[7,8,9]])) temp = T.arange(ones.shape[0]) print(temp.eval())结果 [0 1 2]

import theano import numpy as npones1 = theano.shared(np.float32([[1,2,3],[4,5,6],[7,8,9]])) ones2 = theano.shared(np.float32([[9,8,7],[6,5,4],[3,2,1]]))print(ones1.eval())outputs, updates = theano.sca

import theano import numpy as npones = theano.shared(np.float32([[1,2,3],[4,5,6],[7,8,9]]))print(ones.eval())outputs, updates = theano.scan(lambda result : result + 1, se

https://github.com/MarkWuNLP/MultiTurnResponseSelection/blob/master/src/RNN.py

import theano import numpy as np import theano.tensor as T ones = theano.shared(np.float32([[1,2,3],[4,5,6],[7,8,9]]))print(ones.get_value())result = T.dot(ones,ones)print(result.eval())结果：[[ 1. 2. 3

import theano import numpy as np import theano.tensor as Tones1 = theano.shared(np.float32([[1,2,3],[4,5,6],[7,8,9]]))ones2 = theano.shared(np.float32([[9,8,7],[6,5,4],[3,2,1]]))print(ones1.eval())prin

import theano import numpy as np import theano.tensor as T ones = theano.shared(np.float32([[1,2,3],[4,5,6],[7,8,9]]))print(ones.get_value())result = T.concatenate([ones,ones],axis=0)print(result.eval(

import theano import numpy as np import theano.tensor as T ones = theano.shared(np.float32([[1,2,3],[4,5,6],[7,8,9]]))print(ones.get_value())result = T.stack([ones,ones],axis=0)print(result.eval())resu

import theano import numpy as np import theano.tensor as T numbers = theano.shared(np.float32([[1,2,3],[4,5,6],[7,8,9]])) numbers2 = T.col("temp_col",'float32') temp = numbers+numbers2train = theano.fu