lstm.py

# -*- coding: utf-8 -*-
""" Implement a pyTorch LSTM with hard sigmoid reccurent activation functions.
    Adapted from the non-cuda variant of pyTorch LSTM at
    https://github.com/pytorch/pytorch/blob/master/torch/nn/_functions/rnn.py
"""

from __future__ import print_function, division
import math
import torch

from torch.nn import Module
from torch.nn.parameter import Parameter
from torch.nn.utils.rnn import PackedSequence
import torch.nn.functional as F

class LSTMHardSigmoid(Module):

    def __init__(self, input_size, hidden_size,
                 num_layers=1, bias=True, batch_first=False,
                 dropout=0, bidirectional=False):
        super(LSTMHardSigmoid, self).__init__()
        self.input_size = input_size
        self.hidden_size = hidden_size
        self.num_layers = num_layers
        self.bias = bias
        self.batch_first = batch_first
        self.dropout = dropout
        self.dropout_state = {}
        self.bidirectional = bidirectional
        num_directions = 2 if bidirectional else 1

        gate_size = 4 * hidden_size

        self._all_weights = []
        for layer in range(num_layers):
            for direction in range(num_directions):
                layer_input_size = input_size if layer == 0 else hidden_size * num_directions

                w_ih = Parameter(torch.Tensor(gate_size, layer_input_size))
                w_hh = Parameter(torch.Tensor(gate_size, hidden_size))
                b_ih = Parameter(torch.Tensor(gate_size))
                b_hh = Parameter(torch.Tensor(gate_size))
                layer_params = (w_ih, w_hh, b_ih, b_hh)

                suffix = '_reverse' if direction == 1 else ''
                param_names = ['weight_ih_l{}{}', 'weight_hh_l{}{}']
                if bias:
                    param_names += ['bias_ih_l{}{}', 'bias_hh_l{}{}']
                param_names = [x.format(layer, suffix) for x in param_names]

                for name, param in zip(param_names, layer_params):
                    setattr(self, name, param)
                self._all_weights.append(param_names)

        self.flatten_parameters()
        self.reset_parameters()

    def flatten_parameters(self):
        """Resets parameter data pointer so that they can use faster code paths.

        Right now, this is a no-op wince we don't use CUDA acceleration.
        """
        self._data_ptrs = []

    def _apply(self, fn):
        ret = super(LSTMHardSigmoid, self)._apply(fn)
        self.flatten_parameters()
        return ret

    def reset_parameters(self):
        stdv = 1.0 / math.sqrt(self.hidden_size)
        for weight in self.parameters():
            weight.data.uniform_(-stdv, stdv)

    def forward(self, input, hx=None):
        is_packed = isinstance(input, PackedSequence)
        if is_packed:
            batch_sizes = input.batch_sizes
            input = input.data
            max_batch_size = batch_sizes[0]
        else:
            batch_sizes = None
            max_batch_size = input.size(0) if self.batch_first else input.size(1)

        if hx is None:
            num_directions = 2 if self.bidirectional else 1
            hx = torch.autograd.Variable(input.data.new(self.num_layers *
                                                        num_directions,
                                                        max_batch_size,
                                                        self.hidden_size).zero_(), requires_grad=False)
            hx = (hx, hx)

        has_flat_weights = list(p.data.data_ptr() for p in self.parameters()) == self._data_ptrs
        if has_flat_weights:
            first_data = next(self.parameters()).data
            assert first_data.storage().size() == self._param_buf_size
            flat_weight = first_data.new().set_(first_data.storage(), 0, torch.Size([self._param_buf_size]))
        else:
            flat_weight = None
        func = AutogradRNN(
            self.input_size,
            self.hidden_size,
            num_layers=self.num_layers,
            batch_first=self.batch_first,
            dropout=self.dropout,
            train=self.training,
            bidirectional=self.bidirectional,
            batch_sizes=batch_sizes,
            dropout_state=self.dropout_state,
            flat_weight=flat_weight
        )
        output, hidden = func(input, self.all_weights, hx)
        if is_packed:
            output = PackedSequence(output, batch_sizes)
        return output, hidden

    def __repr__(self):
        s = '{name}({input_size}, {hidden_size}'
        if self.num_layers != 1:
            s += ', num_layers={num_layers}'
        if self.bias is not True:
            s += ', bias={bias}'
        if self.batch_first is not False:
            s += ', batch_first={batch_first}'
        if self.dropout != 0:
            s += ', dropout={dropout}'
        if self.bidirectional is not False:
            s += ', bidirectional={bidirectional}'
        s += ')'
        return s.format(name=self.__class__.__name__, **self.__dict__)

    def __setstate__(self, d):
        super(LSTMHardSigmoid, self).__setstate__(d)
        self.__dict__.setdefault('_data_ptrs', [])
        if 'all_weights' in d:
            self._all_weights = d['all_weights']
        if isinstance(self._all_weights[0][0], str):
            return
        num_layers = self.num_layers
        num_directions = 2 if self.bidirectional else 1
        self._all_weights = []
        for layer in range(num_layers):
            for direction in range(num_directions):
                suffix = '_reverse' if direction == 1 else ''
                weights = ['weight_ih_l{}{}', 'weight_hh_l{}{}', 'bias_ih_l{}{}', 'bias_hh_l{}{}']
                weights = [x.format(layer, suffix) for x in weights]
                if self.bias:
                    self._all_weights += [weights]
                else:
                    self._all_weights += [weights[:2]]

    @property
    def all_weights(self):
        return [[getattr(self, weight) for weight in weights] for weights in self._all_weights]

def AutogradRNN(input_size, hidden_size, num_layers=1, batch_first=False,
                dropout=0, train=True, bidirectional=False, batch_sizes=None,
                dropout_state=None, flat_weight=None):

    cell = LSTMCell

    if batch_sizes is None:
        rec_factory = Recurrent
    else:
        rec_factory = variable_recurrent_factory(batch_sizes)

    if bidirectional:
        layer = (rec_factory(cell), rec_factory(cell, reverse=True))
    else:
        layer = (rec_factory(cell),)

    func = StackedRNN(layer,
                      num_layers,
                      True,
                      dropout=dropout,
                      train=train)

    def forward(input, weight, hidden):
        if batch_first and batch_sizes is None:
            input = input.transpose(0, 1)

        nexth, output = func(input, hidden, weight)

        if batch_first and batch_sizes is None:
            output = output.transpose(0, 1)

        return output, nexth

    return forward

def Recurrent(inner, reverse=False):
    def forward(input, hidden, weight):
        output = []
        steps = range(input.size(0) - 1, -1, -1) if reverse else range(input.size(0))
        for i in steps:
            hidden = inner(input[i], hidden, *weight)
            # hack to handle LSTM
            output.append(hidden[0] if isinstance(hidden, tuple) else hidden)

        if reverse:
            output.reverse()
        output = torch.cat(output, 0).view(input.size(0), *output[0].size())

        return hidden, output

    return forward


def variable_recurrent_factory(batch_sizes):
    def fac(inner, reverse=False):
        if reverse:
            return VariableRecurrentReverse(batch_sizes, inner)
        else:
            return VariableRecurrent(batch_sizes, inner)
    return fac

def VariableRecurrent(batch_sizes, inner):
    def forward(input, hidden, weight):
        output = []
        input_offset = 0
        last_batch_size = batch_sizes[0]
        hiddens = []
        flat_hidden = not isinstance(hidden, tuple)
        if flat_hidden:
            hidden = (hidden,)
        for batch_size in batch_sizes:
            step_input = input[input_offset:input_offset + batch_size]
            input_offset += batch_size

            dec = last_batch_size - batch_size
            if dec > 0:
                hiddens.append(tuple(h[-dec:] for h in hidden))
                hidden = tuple(h[:-dec] for h in hidden)
            last_batch_size = batch_size

            if flat_hidden:
                hidden = (inner(step_input, hidden[0], *weight),)
            else:
                hidden = inner(step_input, hidden, *weight)

            output.append(hidden[0])
        hiddens.append(hidden)
        hiddens.reverse()

        hidden = tuple(torch.cat(h, 0) for h in zip(*hiddens))
        assert hidden[0].size(0) == batch_sizes[0]
        if flat_hidden:
            hidden = hidden[0]
        output = torch.cat(output, 0)

        return hidden, output

    return forward


def VariableRecurrentReverse(batch_sizes, inner):
    def forward(input, hidden, weight):
        output = []
        input_offset = input.size(0)
        last_batch_size = batch_sizes[-1]
        initial_hidden = hidden
        flat_hidden = not isinstance(hidden, tuple)
        if flat_hidden:
            hidden = (hidden,)
            initial_hidden = (initial_hidden,)
        hidden = tuple(h[:batch_sizes[-1]] for h in hidden)
        for batch_size in reversed(batch_sizes):
            inc = batch_size - last_batch_size
            if inc > 0:
                hidden = tuple(torch.cat((h, ih[last_batch_size:batch_size]), 0)
                               for h, ih in zip(hidden, initial_hidden))
            last_batch_size = batch_size
            step_input = input[input_offset - batch_size:input_offset]
            input_offset -= batch_size

            if flat_hidden:
                hidden = (inner(step_input, hidden[0], *weight),)
            else:
                hidden = inner(step_input, hidden, *weight)
            output.append(hidden[0])

        output.reverse()
        output = torch.cat(output, 0)
        if flat_hidden:
            hidden = hidden[0]
        return hidden, output

    return forward

def StackedRNN(inners, num_layers, lstm=False, dropout=0, train=True):

    num_directions = len(inners)
    total_layers = num_layers * num_directions

    def forward(input, hidden, weight):
        assert(len(weight) == total_layers)
        next_hidden = []

        if lstm:
            hidden = list(zip(*hidden))

        for i in range(num_layers):
            all_output = []
            for j, inner in enumerate(inners):
                l = i * num_directions + j

                hy, output = inner(input, hidden[l], weight[l])
                next_hidden.append(hy)
                all_output.append(output)

            input = torch.cat(all_output, input.dim() - 1)

            if dropout != 0 and i < num_layers - 1:
                input = F.dropout(input, p=dropout, training=train, inplace=False)

        if lstm:
            next_h, next_c = zip(*next_hidden)
            next_hidden = (
                torch.cat(next_h, 0).view(total_layers, *next_h[0].size()),
                torch.cat(next_c, 0).view(total_layers, *next_c[0].size())
            )
        else:
            next_hidden = torch.cat(next_hidden, 0).view(
                total_layers, *next_hidden[0].size())

        return next_hidden, input

    return forward

def LSTMCell(input, hidden, w_ih, w_hh, b_ih=None, b_hh=None):
    """
    A modified LSTM cell with hard sigmoid activation on the input, forget and output gates.
    """
    hx, cx = hidden
    gates = F.linear(input, w_ih, b_ih) + F.linear(hx, w_hh, b_hh)

    ingate, forgetgate, cellgate, outgate = gates.chunk(4, 1)

    ingate = hard_sigmoid(ingate)
    forgetgate = hard_sigmoid(forgetgate)
    cellgate = F.tanh(cellgate)
    outgate = hard_sigmoid(outgate)

    cy = (forgetgate * cx) + (ingate * cellgate)
    hy = outgate * F.tanh(cy)

    return hy, cy

def hard_sigmoid(x):
    """
    Computes element-wise hard sigmoid of x.
    See e.g. https://github.com/Theano/Theano/blob/master/theano/tensor/nnet/sigm.py#L279
    """
    x = (0.2 * x) + 0.5
    x = F.threshold(-x, -1, -1)
    x = F.threshold(-x, 0, 0)
    return x