rnn.py

import warnings
from torch.autograd import Function, NestedIOFunction, Variable
import torch.backends.cudnn as cudnn
from .. import functional as F
from .thnn import rnnFusedPointwise as fusedBackend

try:
    import torch.backends.cudnn.rnn
except ImportError:
    pass


def RNNReLUCell(input, hidden, w_ih, w_hh, b_ih=None, b_hh=None):
    hy = F.relu(F.linear(input, w_ih, b_ih) + F.linear(hidden, w_hh, b_hh))
    return hy


def RNNTanhCell(input, hidden, w_ih, w_hh, b_ih=None, b_hh=None):
    hy = F.tanh(F.linear(input, w_ih, b_ih) + F.linear(hidden, w_hh, b_hh))
    return hy


def LSTMCell(input, hidden, w_ih, w_hh, b_ih=None, b_hh=None):
    if input.is_cuda:
        igates = F.linear(input, w_ih)
        hgates = F.linear(hidden[0], w_hh)
        state = fusedBackend.LSTMFused()
        return state(igates, hgates, hidden[1]) if b_ih is None else state(igates, hgates, hidden[1], b_ih, b_hh)

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

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

    return hy, cy


def LSTMCell_AddC(input, hidden, w_ih, w_hh, w_ch, b_ih=None, b_hh=None, b_ch=None):
    #print("input : ",input.size())
    #print("hidden_hx : ",hidden[0].size())
    #print("hidden_cx : ",hidden[1].size())
    #print("w_ih : ",w_ih.size())
    #print("w_hh : ",w_hh.size())
    #print("w_ch : ",w_ch.size())
    #assert input.size(0)==5
    #assert input.size(1)==10
    w_ih = w_ih
    w_hh = w_hh
    w_ch = w_ch
    hx, cx = hidden
    #assert w_ih.size(0)==10
    #assert w_ih.size(1)==5

    #assert w_hh.size(0)==5
    #assert w_hh.size(1)==5

    #assert w_ch.size(0)==5
    #assert w_ch.size(1)==5


    gates = F.linear(input, w_ih.t(), b_ih) + F.linear(hx, w_hh.t(), b_hh) #(5,20)+(5,20)

    #print((F.linear(cx, w_ch.t(), b_ch)).size())
    gates_c = gates + F.linear(cx, w_ch.t(), b_ch)#(5,20)

    ingate, forgetgate, cellgate, outgate = gates.chunk(4, 1)
    ingate_, forgetgate_, cellgate_, output_ = gates_c.chunk(4, 1)

    ingate = F.sigmoid(ingate_)
    forgetgate = F.sigmoid(forgetgate_)
    cellgate = F.tanh(cellgate)

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


    return hy, cy



def LSTMCell_AddC_decoder(input, hidden, w_ih, w_hh,w_ch,w_oh, w_oo,  b_ih=None, b_hh=None, b_ch=None, b_oh=None, b_oo = None):
    #print("input : ",input.size())
    #print("hidden_hx : ",hidden[0].size())
    #print("hidden_cx : ",hidden[1].size())
    #print("w_ih : ",w_ih.size())
    #print("w_hh : ",w_hh.size())
    #print("w_ch : ",w_ch.size())
    #print("w_oh : ",w_oh.size())
    #print("w_oo : ",w_oo.size())
    #assert input.size(0)==5
    #assert input.size(1)==10
    w_ih = w_ih
    w_hh = w_hh
    w_ch = w_ch
    w_oh = w_oh
    hx, cx, ox = hidden
    #assert w_ih.size(0)==10
    #assert w_ih.size(1)==5

    #assert w_hh.size(0)==5
    #assert w_hh.size(1)==5

    #assert w_ch.size(0)==5
    #assert w_ch.size(1)==5
    gates = F.linear(input, w_ih.t(),b_ih) + F.linear(hx, w_hh.t(),b_hh) #[15,300] * [300,1200]
    gates_o = gates + F.linear(ox, w_oh.t(), b_oh)#(5,20)
    ingate, forgetgate, cellgate, outgate = gates.chunk(4, 1)
    ingate_, forgetgate_, cellgate_, output_ = gates_o.chunk(4, 1)

    ingate = F.sigmoid(ingate_)
    forgetgate = F.sigmoid(forgetgate_)
    cellgate = F.tanh(cellgate_)

    cy = (forgetgate * cx) + (ingate * cellgate) #(5,5) * (5,5) + (5,5)*(5,5)
    outgate = outgate + cy
    outgate = F.sigmoid(outgate)
    hy = outgate * F.tanh(cy)
    oy = F.linear(hy, w_oo.t(), b_oo) #(5,5) * (5,5)

    return hy, cy, oy


def GRUCell(input, hidden, w_ih, w_hh, b_ih=None, b_hh=None):
    if input.is_cuda:
        gi = F.linear(input, w_ih)
        gh = F.linear(hidden, w_hh)
        state = fusedBackend.GRUFused()
        return state(gi, gh, hidden) if b_ih is None else state(gi, gh, hidden, b_ih, b_hh)

    gi = F.linear(input, w_ih, b_ih)
    gh = F.linear(hidden, w_hh, b_hh)
    i_r, i_i, i_n = gi.chunk(3, 1)
    h_r, h_i, h_n = gh.chunk(3, 1)

    resetgate = F.sigmoid(i_r + h_r)
    inputgate = F.sigmoid(i_i + h_i)
    newgate = F.tanh(i_n + resetgate * h_n)
    hy = newgate + inputgate * (hidden - newgate)

    return hy


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

            #input = torch.cat(all_output, input.dim() - 1)
            input = torch.add(*all_output)
            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 StackedRNN_addc_decoder(inners, num_layers, lstm=False, dropout=0, train=True):
    num_directions = len(inners)
    total_layers = num_layers * num_directions

    def forward(input, hidden, weight):
        #import ipdb;ipdb.set_trace();
        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)
            #input = torch.add(*all_output) # 1 * 15 * 300
            if dropout != 0 and i < num_layers - 1:
                input = F.dropout(input, p=dropout, training=train, inplace=False)

        if lstm:
            next_h, next_c, next_o = 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()),
                torch.cat(next_o, 0).view(total_layers, *next_o[0].size())
            )
        else:
            #import ipdb; ipdb.set_trace();
            next_hidden = torch.cat(next_hidden, 0).view(
                total_layers, *next_hidden[0].size())

        return next_hidden, input

    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())

        #print("current hidden : ",hidden)
        #print("current output : ",output)
        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 AutogradRNN(mode, 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,procedure='encoder'):
    if mode == 'RNN_RELU':
        cell = RNNReLUCell
    elif mode == 'RNN_TANH':
        cell = RNNTanhCell
    elif mode == 'LSTM':
        cell = LSTMCell
    elif mode == 'GRU':
        cell = GRUCell
    elif mode == 'LSTMCell_AddC':
        cell = LSTMCell_AddC
    elif mode == 'LSTMCell_AddC_decoder':
        cell = LSTMCell_AddC_decoder
    else:
        raise Exception('Unknown mode: {}'.format(mode))

    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),)


    if(procedure == 'encoder'):
        func = StackedRNN(layer,
                          num_layers,
                          lstm=("LSTM" in mode),
                          dropout=dropout,
                          train=train)
    else:
        func = StackedRNN_addc_decoder(layer,
                                       num_layers,
                                       lstm=("LSTM" in mode),
                                       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


class CudnnRNN(NestedIOFunction):
    def __init__(self, mode, 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):
        super(CudnnRNN, self).__init__()
        if dropout_state is None:
            dropout_state = {}
        self.mode = cudnn.rnn.get_cudnn_mode(mode)
        self.input_mode = cudnn.CUDNN_LINEAR_INPUT
        self.input_size = input_size
        self.hidden_size = hidden_size
        self.num_layers = num_layers
        self.batch_first = batch_first
        self.dropout = dropout
        self.train = train
        self.bidirectional = 1 if bidirectional else 0
        self.num_directions = 2 if bidirectional else 1
        self.batch_sizes = batch_sizes
        self.dropout_seed = torch.IntTensor(1).random_()[0]
        self.dropout_state = dropout_state
        self.weight_buf = flat_weight
        if flat_weight is None:
            warnings.warn("RNN module weights are not part of single contiguous "
                          "chunk of memory. This means they need to be compacted "
                          "at every call, possibly greately increasing memory usage. "
                          "To compact weights again call flatten_parameters().", stacklevel=5)

    def forward_extended(self, input, weight, hx):
        assert cudnn.is_acceptable(input)
        # TODO: raise a warning if weight_data_ptr is None

        output = input.new()

        if torch.is_tensor(hx):
            hy = hx.new()
        else:
            hy = tuple(h.new() for h in hx)

        cudnn.rnn.forward(self, input, hx, weight, output, hy)

        self.save_for_backward(input, hx, weight, output)
        return output, hy

    def backward_extended(self, grad_output, grad_hy):
        input, hx, weight, output = self.saved_tensors
        input = input.contiguous()

        grad_input, grad_weight, grad_hx = None, None, None

        assert cudnn.is_acceptable(input)

        grad_input = input.new()
        if torch.is_tensor(hx):
            grad_hx = input.new()
        else:
            grad_hx = tuple(h.new() for h in hx)

        if self.retain_variables:
            self._reserve_clone = self.reserve.clone()

        cudnn.rnn.backward_grad(
            self,
            input,
            hx,
            weight,
            output,
            grad_output,
            grad_hy,
            grad_input,
            grad_hx)

        if any(self.needs_input_grad[1:]):
            grad_weight = [tuple(w.new().resize_as_(w) for w in layer_weight) for layer_weight in weight]
            cudnn.rnn.backward_weight(
                self,
                input,
                hx,
                output,
                weight,
                grad_weight)
        else:
            grad_weight = [(None,) * len(layer_weight) for layer_weight in weight]

        if self.retain_variables:
            self.reserve = self._reserve_clone
            del self._reserve_clone

        return grad_input, grad_weight, grad_hx


def RNN(*args, **kwargs):
    def forward(input, *fargs, **fkwargs):
        if cudnn.is_acceptable(input.data):
            func = CudnnRNN(*args, **kwargs)
        else:
            func = AutogradRNN(*args, **kwargs)
        return func(input, *fargs, **fkwargs)

    return forward