RecurrentLayer.compute_tensor() - Code Metrics - Inspection of "Changed dataset" - zomux/deepy - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( f73e69...91b7c0 )

by Raphael

created 2016-05-23 15:37 UTC

RecurrentLayer.compute_tensor() F

↳ Parent: RecurrentLayer

Complexity

Conditions

Size

Total Lines

Duplication

Lines	0
Ratio	0 %

Metric	Value
cc	10
dl	0
loc	35
rs	3.1304

How to fix Complexity

#!/usr/bin/env python
# -*- coding: utf-8 -*-

from . import NeuralLayer
from var import NeuralVariable
from deepy.utils import build_activation, FLOATX, XavierGlorotInitializer, OrthogonalInitializer, Scanner, neural_computation
import numpy as np
import theano.tensor as T
from abc import ABCMeta, abstractmethod

OUTPUT_TYPES = ["sequence", "one"]
INPUT_TYPES = ["sequence", "one"]



class RecurrentLayer(NeuralLayer):
    __metaclass__ = ABCMeta

    def __init__(self, name, state_names, hidden_size=100, input_type="sequence", output_type="sequence",
                 inner_init=None, outer_init=None,
                 gate_activation='sigmoid', activation='tanh',
                 steps=None, backward=False, mask=None,
                 additional_input_dims=None):
        super(RecurrentLayer, self).__init__(name)
        self.state_names = state_names
        self.main_state = state_names[0]
        self.hidden_size = hidden_size
        self._gate_activation = gate_activation
        self._activation = activation
        self.gate_activate = build_activation(self._gate_activation)
        self.activate = build_activation(self._activation)
        self._input_type = input_type
        self._output_type = output_type
        self.inner_init = inner_init if inner_init else OrthogonalInitializer()
        self.outer_init = outer_init if outer_init else XavierGlorotInitializer()
        self._steps = steps
        self._mask = mask.tensor if type(mask) == NeuralVariable else mask
        self._go_backwards = backward
        self.additional_input_dims = additional_input_dims if additional_input_dims else []

        if input_type not in INPUT_TYPES:
            raise Exception("Input type of {} is wrong: {}".format(name, input_type))
        if output_type not in OUTPUT_TYPES:
            raise Exception("Output type of {} is wrong: {}".format(name, output_type))

    @neural_computation
    def step(self, step_inputs):
        new_states = self.compute_new_state(step_inputs)

        # apply mask for each step if `output_type` is 'one'
        if self._output_type == "one" and step_inputs.get("mask"):
            mask = step_inputs["mask"].dimshuffle(0, 'x')
            for state_name in new_states:
                new_states[state_name] = new_states[state_name] * mask + step_inputs[state_name] * (1 - mask)

        return new_states

    @abstractmethod
    def compute_new_state(self, step_inputs):
        """
        :type step_inputs: dict
        :rtype: dict
        """

    @abstractmethod
    def merge_inputs(self, input_var, additional_inputs=None):
        """
        Merge inputs and return a map, which will be passed to core_step.
        :type input_var: T.var
        :param additional_inputs: list
        :rtype: dict
        """

    @abstractmethod
    def prepare(self):
        pass

    @neural_computation
    def get_initial_states(self, input_var):
        """
        :type input_var: T.var
        :rtype: dict
        """
        initial_states = {}
        for state in self.state_names:
            initial_states[state] = T.alloc(np.cast[FLOATX](0.), input_var.shape[0], self.hidden_size)
        return initial_states

    @neural_computation
    def get_step_inputs(self, input_var, states=None, mask=None, additional_inputs=None):
        """
        :type input_var: T.var
        :rtype: dict
        """
        step_inputs = {}
        if self._input_type == "sequence":
            if not additional_inputs:
                additional_inputs = []
            step_inputs.update(self.merge_inputs(input_var, additional_inputs=additional_inputs))
        else:
            # step_inputs["mask"] = mask.dimshuffle((1,0)) if mask else None
            if additional_inputs:
                step_inputs.update(self.merge_inputs(None, additional_inputs=additional_inputs))
        if states:
            for name in self.state_names:
                step_inputs[name] = states[name]

        return step_inputs

    def compute(self, input_var, mask=None, additional_inputs=None, steps=None, backward=False):
        if additional_inputs and not self.additional_input_dims:
            self.additional_input_dims = map(lambda var: var.dim(), additional_inputs)
        return super(RecurrentLayer, self).compute(input_var, mask=mask, additional_inputs=additional_inputs, steps=steps, backward=backward)

    def compute_tensor(self, input_var, mask=None, additional_inputs=None, steps=None, backward=False):
        # prepare parameters
        backward = backward if backward else self._go_backwards
        steps = steps if steps else self._steps
        mask = mask if mask else self._mask
        if mask and self._input_type == "one":
            raise Exception("Mask only works with sequence input")
        # get initial states
        init_state_map = self.get_initial_states(input_var)
        # get input sequence map
        if self._input_type == "sequence":
            # Move middle dimension to left-most position
            # (sequence, batch, value)
            input_var = input_var.dimshuffle((1,0,2))
            seq_map = self.get_step_inputs(input_var, mask=mask, additional_inputs=additional_inputs)
        else:
            init_state_map[self.main_state] = input_var
            seq_map = self.get_step_inputs(None, mask=mask, additional_inputs=additional_inputs)
        # scan
        retval_map, _ = Scanner(
            self.step,
            sequences=seq_map,
            outputs_info=init_state_map,
            n_steps=steps,
            go_backwards=backward
        ).compute()
        # return main states
        main_states = retval_map[self.main_state]
        if self._output_type == "one":
            return main_states[-1]
        elif self._output_type == "sequence":
            main_states = main_states.dimshuffle((1,0,2)) # ~ batch, time, size
            if mask: # ~ batch, time
                main_states *= mask.dimshuffle((0, 1, 'x'))
            return main_states


class RNN(RecurrentLayer):

    def  __init__(self, hidden_size, **kwargs):
        kwargs["hidden_size"] = hidden_size
        super(RNN, self).__init__("RNN", ["state"], **kwargs)

    @neural_computation
    def compute_new_state(self, step_inputs):
        xh_t, h_tm1 = map(step_inputs.get, ["xh_t", "state"])
        if not xh_t:
            xh_t = 0

        h_t = self.activate(xh_t + T.dot(h_tm1, self.W_h) + self.b_h)

        return {"state": h_t}

    @neural_computation
    def merge_inputs(self, input_var, additional_inputs=None):
        if not additional_inputs:
            additional_inputs = []
        all_inputs = [input_var] + additional_inputs
        h_inputs = []
        for x, weights in zip(all_inputs, self.input_weights):
            wi, = weights
            h_inputs.append(T.dot(x, wi))
        merged_inputs = {
            "xh_t": sum(h_inputs)
        }
        return merged_inputs

    def prepare(self):
        self.output_dim = self.hidden_size

        self.W_h = self.create_weight(self.hidden_size, self.hidden_size, "h", initializer=self.outer_init)
        self.b_h = self.create_bias(self.hidden_size, "h")

        self.register_parameters(self.W_h, self.b_h)

        self.input_weights = []
        if self._input_type == "sequence":
            all_input_dims = [self.input_dim] + self.additional_input_dims
            for i, input_dim in enumerate(all_input_dims):
                wi = self.create_weight(input_dim, self.hidden_size, "wi_{}".format(i+1), initializer=self.outer_init)
                weights = [wi]
                self.input_weights.append(weights)
                self.register_parameters(*weights)

1			#!/usr/bin/env python
2			# -- coding: utf-8 --
3
4			from . import NeuralLayer
5			from var import NeuralVariable
6			from deepy.utils import build_activation, FLOATX, XavierGlorotInitializer, OrthogonalInitializer, Scanner, neural_computation
7			import numpy as np
8			import theano.tensor as T
9			from abc import ABCMeta, abstractmethod
10
11			OUTPUT_TYPES = ["sequence", "one"]
12			INPUT_TYPES = ["sequence", "one"]
13
14
15
16			class RecurrentLayer(NeuralLayer):
17			__metaclass__ = ABCMeta
18
19			def __init__(self, name, state_names, hidden_size=100, input_type="sequence", output_type="sequence",
20			inner_init=None, outer_init=None,
21			gate_activation='sigmoid', activation='tanh',
22			steps=None, backward=False, mask=None,
23			additional_input_dims=None):
24			super(RecurrentLayer, self).__init__(name)
25			self.state_names = state_names
26			self.main_state = state_names[0]
27			self.hidden_size = hidden_size
28			self._gate_activation = gate_activation
29			self._activation = activation
30			self.gate_activate = build_activation(self._gate_activation)
31			self.activate = build_activation(self._activation)
32			self._input_type = input_type
33			self._output_type = output_type
34			self.inner_init = inner_init if inner_init else OrthogonalInitializer()
35			self.outer_init = outer_init if outer_init else XavierGlorotInitializer()
36			self._steps = steps
37			self._mask = mask.tensor if type(mask) == NeuralVariable else mask
38			self._go_backwards = backward
39			self.additional_input_dims = additional_input_dims if additional_input_dims else []
40
41			if input_type not in INPUT_TYPES:
42			raise Exception("Input type of {} is wrong: {}".format(name, input_type))
43			if output_type not in OUTPUT_TYPES:
44			raise Exception("Output type of {} is wrong: {}".format(name, output_type))
45
46			@neural_computation
47			def step(self, step_inputs):
48			new_states = self.compute_new_state(step_inputs)
49
50			# apply mask for each step if `output_type` is 'one'
51			if self._output_type == "one" and step_inputs.get("mask"):
52			mask = step_inputs["mask"].dimshuffle(0, 'x')
53			for state_name in new_states:
54			new_states[state_name] = new_states[state_name] * mask + step_inputs[state_name] * (1 - mask)
55
56			return new_states
57
58			@abstractmethod
59			def compute_new_state(self, step_inputs):
60			"""
61			:type step_inputs: dict
62			:rtype: dict
63			"""
64
65			@abstractmethod
66			def merge_inputs(self, input_var, additional_inputs=None):
67			"""
68			Merge inputs and return a map, which will be passed to core_step.
69			:type input_var: T.var
70			:param additional_inputs: list
71			:rtype: dict
72			"""
73
74			@abstractmethod
75			def prepare(self):
76			pass
77
78			@neural_computation
79			def get_initial_states(self, input_var):
80			"""
81			:type input_var: T.var
82			:rtype: dict
83			"""
84			initial_states = {}
85			for state in self.state_names:
86			initial_states[state] = T.alloc(np.cast[FLOATX](0.), input_var.shape[0], self.hidden_size)
87			return initial_states
88
89			@neural_computation
90			def get_step_inputs(self, input_var, states=None, mask=None, additional_inputs=None):
91			"""
92			:type input_var: T.var
93			:rtype: dict
94			"""
95			step_inputs = {}
96			if self._input_type == "sequence":
97			if not additional_inputs:
98			additional_inputs = []
99			step_inputs.update(self.merge_inputs(input_var, additional_inputs=additional_inputs))
100			else:
101			# step_inputs["mask"] = mask.dimshuffle((1,0)) if mask else None
102			if additional_inputs:
103			step_inputs.update(self.merge_inputs(None, additional_inputs=additional_inputs))
104			if states:
105			for name in self.state_names:
106			step_inputs[name] = states[name]
107
108			return step_inputs
109
110			def compute(self, input_var, mask=None, additional_inputs=None, steps=None, backward=False):
111			if additional_inputs and not self.additional_input_dims:
112			self.additional_input_dims = map(lambda var: var.dim(), additional_inputs)
113			return super(RecurrentLayer, self).compute(input_var, mask=mask, additional_inputs=additional_inputs, steps=steps, backward=backward)
114
115			def compute_tensor(self, input_var, mask=None, additional_inputs=None, steps=None, backward=False):
116			# prepare parameters
117			backward = backward if backward else self._go_backwards
118			steps = steps if steps else self._steps
119			mask = mask if mask else self._mask
120			if mask and self._input_type == "one":
121			raise Exception("Mask only works with sequence input")
122			# get initial states
123			init_state_map = self.get_initial_states(input_var)
124			# get input sequence map
125			if self._input_type == "sequence":
126			# Move middle dimension to left-most position
127			# (sequence, batch, value)
128			input_var = input_var.dimshuffle((1,0,2))
129			seq_map = self.get_step_inputs(input_var, mask=mask, additional_inputs=additional_inputs)
130			else:
131			init_state_map[self.main_state] = input_var
132			seq_map = self.get_step_inputs(None, mask=mask, additional_inputs=additional_inputs)
133			# scan
134			retval_map, _ = Scanner(
135			self.step,
136			sequences=seq_map,
137			outputs_info=init_state_map,
138			n_steps=steps,
139			go_backwards=backward
140			).compute()
141			# return main states
142			main_states = retval_map[self.main_state]
143			if self._output_type == "one":
144			return main_states[-1]
145			elif self._output_type == "sequence":
146			main_states = main_states.dimshuffle((1,0,2)) # ~ batch, time, size
147			if mask: # ~ batch, time
148			main_states *= mask.dimshuffle((0, 1, 'x'))
149			return main_states
150
151
152			class RNN(RecurrentLayer):
153
154			def __init__(self, hidden_size, **kwargs):
155			kwargs["hidden_size"] = hidden_size
156			super(RNN, self).__init__("RNN", ["state"], **kwargs)
157
158			@neural_computation
159			def compute_new_state(self, step_inputs):
160			xh_t, h_tm1 = map(step_inputs.get, ["xh_t", "state"])
161			if not xh_t:
162			xh_t = 0
163
164			h_t = self.activate(xh_t + T.dot(h_tm1, self.W_h) + self.b_h)
165
166			return {"state": h_t}
167
168			@neural_computation
169			def merge_inputs(self, input_var, additional_inputs=None):
170			if not additional_inputs:
171			additional_inputs = []
172			all_inputs = [input_var] + additional_inputs
173			h_inputs = []
174			for x, weights in zip(all_inputs, self.input_weights):
175			wi, = weights
176			h_inputs.append(T.dot(x, wi))
177			merged_inputs = {
178			"xh_t": sum(h_inputs)
179			}
180			return merged_inputs
181
182			def prepare(self):
183			self.output_dim = self.hidden_size
184
185			self.W_h = self.create_weight(self.hidden_size, self.hidden_size, "h", initializer=self.outer_init)
186			self.b_h = self.create_bias(self.hidden_size, "h")
187
188			self.register_parameters(self.W_h, self.b_h)
189
190			self.input_weights = []
191			if self._input_type == "sequence":
192			all_input_dims = [self.input_dim] + self.additional_input_dims
193			for i, input_dim in enumerate(all_input_dims):
194			wi = self.create_weight(input_dim, self.hidden_size, "wi_{}".format(i+1), initializer=self.outer_init)
195			weights = [wi]
196			self.input_weights.append(weights)
197			self.register_parameters(*weights)

zomux / deepy

Push — master ( f73e69...91b7c0 )

RecurrentLayer.compute_tensor() F

Complexity

Size

Duplication

How to fix Complexity

Complexity

Duplication Side-by-Side

Filter issues like