el commit inicial

Author: zygmuntz

PointerLSTM.py

@@ -0,0 +1,64 @@
+from keras import initializations
+from keras.layers.recurrent import time_distributed_dense
+from keras.activations import tanh, softmax
+from keras.layers import LSTM
+from keras.engine import InputSpec
+import keras.backend as K
+
+
+class PointerLSTM(LSTM):
+    def __init__(self, hidden_shape, *args, **kwargs):
+        self.hidden_shape = hidden_shape
+        self.input_length = []
+        super(PointerLSTM, self).__init__(*args, **kwargs)
+
+    def build(self, input_shape):
+        super(PointerLSTM, self).build(input_shape)
+        self.input_spec = [InputSpec(shape=input_shape)]
+        init = initializations.get('orthogonal')
+        self.W1 = init((self.hidden_shape, 1))
+        self.W2 = init((self.hidden_shape, 1))
+        self.vt = init((input_shape[1], 1))
+        self.trainable_weights += [self.W1, self.W2, self.vt]
+
+    def call(self, x, mask=None):
+        input_shape = self.input_spec[0].shape
+        en_seq = x
+        x_input = x[:, input_shape[1]-1, :]
+        x_input = K.repeat(x_input, input_shape[1])
+        initial_states = self.get_initial_states(x_input)
+
+        constants = super(PointerLSTM, self).get_constants(x_input)
+        constants.append(en_seq)
+        preprocessed_input = self.preprocess_input(x_input)
+
+        last_output, outputs, states = K.rnn(self.step, preprocessed_input,
+                                             initial_states,
+                                             go_backwards=self.go_backwards,
+                                             constants=constants,
+                                             input_length=input_shape[1])
+
+        return outputs
+
+    def step(self, x_input, states):
+    	#print "x_input:", x_input, x_input.shape
+    	# <TensorType(float32, matrix)>
+    	
+        input_shape = self.input_spec[0].shape
+        en_seq = states[-1]
+        _, [h, c] = super(PointerLSTM, self).step(x_input, states[:-1])
+
+        # vt*tanh(W1*e+W2*d)
+        dec_seq = K.repeat(h, input_shape[1])
+        Eij = time_distributed_dense(en_seq, self.W1, output_dim=1)
+        Dij = time_distributed_dense(dec_seq, self.W2, output_dim=1)
+        U = self.vt * tanh(Eij + Dij)
+        U = K.squeeze(U, 2)
+
+        # make probability tensor
+        pointer = softmax(U)
+        return pointer, [h, c]
+
+    def get_output_shape_for(self, input_shape):
+        # output shape is not affected by the attention component
+        return (input_shape[0], input_shape[1], input_shape[1])

README.md

@@ -1 +1,7 @@
-# pointer-networks-experiments
+# Pointer networks experiments
+
+Code for genereting data, training and testing. There are two tasks: ordering single numbers, or sums or numbers (the `_sums` suffix).
+
+The training scripts save model weights each epoch. When started, they attempt to load appropriate weights - corresponding to the hidden size and the chosen number of steps in a sequence.
+
+This software builts upon [https://github.com/keon/pointer-networks](https://github.com/keon/pointer-networks).