import torch import torch.nn as nn from torch.autograd import Variable from torch import optim import torch.nn.functional as F import math, copy, time import torch.nn.utils.rnn as rnn_utils from data import get_cuda, to_var, calc_bleu import numpy as np def clones(module, N): """Produce N identical layers.""" return nn.ModuleList([copy.deepcopy(module) for _ in range(N)]) def attention(query, key, value, mask=None, dropout=None): """Compute 'Scaled Dot Product Attention' """ d_k = query.size(-1) scores = torch.matmul(query, key.transpose(-2, -1)) / math.sqrt(d_k) if mask is not None: scores = scores.masked_fill(mask == 0, -1e9) p_attn = F.softmax(scores, dim=-1) if dropout is not None: p_attn = dropout(p_attn) return torch.matmul(p_attn, value), p_attn class MultiHeadedAttention(nn.Module): def __init__(self, h, d_model, dropout=0.1): """Take in model size and number of heads.""" super(MultiHeadedAttention, self).__init__() assert d_model % h == 0 # We assume d_v always equals d_k self.d_k = d_model // h self.h = h self.linears = clones(nn.Linear(d_model, d_model), 4) self.attn = None self.dropout = nn.Dropout(p=dropout) def forward(self, query, key, value, mask=None): if mask is not None: # Same mask applied to all h heads. mask = mask.unsqueeze(1) nbatches = query.size(0) # 1) Do all the linear projections in batch from d_model => h x d_k query, key, value = \ [l(x).view(nbatches, -1, self.h, self.d_k).transpose(1, 2) for l, x in zip(self.linears, (query, key, value))] # 2) Apply attention on all the projected vectors in batch. x, self.attn = attention(query, key, value, mask=mask, dropout=self.dropout) # 3) "Concat" using a view and apply a final linear. x = x.transpose(1, 2).contiguous() \ .view(nbatches, -1, self.h * self.d_k) return self.linears[-1](x) class PositionwiseFeedForward(nn.Module): """Implements FFN equation.""" def __init__(self, d_model, d_ff, dropout=0.1): super(PositionwiseFeedForward, self).__init__() self.w_1 = nn.Linear(d_model, d_ff) self.w_2 = nn.Linear(d_ff, d_model) self.dropout = nn.Dropout(dropout) def forward(self, x): return self.w_2(self.dropout(F.relu(self.w_1(x)))) class PositionalEncoding(nn.Module): """Implement the PE function.""" def __init__(self, d_model, dropout, max_len=5000): super(PositionalEncoding, self).__init__() self.dropout = nn.Dropout(p=dropout) # Compute the positional encodings once in log space. pe = torch.zeros(max_len, d_model) position = torch.arange(0, max_len).unsqueeze(1) div_term = torch.exp(torch.arange(0, d_model, 2) * -(math.log(10000.0) / d_model)) pe[:, 0::2] = torch.sin(position * div_term) pe[:, 1::2] = torch.cos(position * div_term) pe = pe.unsqueeze(0) self.register_buffer('pe', pe) def forward(self, x): x = x + Variable(self.pe[:, :x.size(1)], requires_grad=False) return self.dropout(x) class LayerNorm(nn.Module): """Construct a layernorm module (See citation for details).""" def __init__(self, features, eps=1e-6): super(LayerNorm, self).__init__() self.a_2 = nn.Parameter(torch.ones(features)) self.b_2 = nn.Parameter(torch.zeros(features)) self.eps = eps def forward(self, x): mean = x.mean(-1, keepdim=True) std = x.std(-1, keepdim=True) return self.a_2 * (x - mean) / (std + self.eps) + self.b_2 class SublayerConnection(nn.Module): """ A residual connection followed by a layer norm. Note for code simplicity the norm is first as opposed to last. """ def __init__(self, size, dropout): super(SublayerConnection, self).__init__() self.norm = LayerNorm(size) self.dropout = nn.Dropout(dropout) def forward(self, x, sublayer): """Apply residual connection to any sublayer with the same size.""" return x + self.dropout(sublayer(self.norm(x))) class Embeddings(nn.Module): def __init__(self, d_model, vocab): super(Embeddings, self).__init__() self.lut = nn.Embedding(vocab, d_model) self.d_model = d_model def forward(self, x): return self.lut(x) * math.sqrt(self.d_model) ################ Encoder ################ class Encoder(nn.Module): """Core encoder is a stack of N layers""" def __init__(self, layer, N): super(Encoder, self).__init__() self.layers = clones(layer, N) self.norm = LayerNorm(layer.size) def forward(self, x, mask): """Pass the input (and mask) through each layer in turn.""" for layer in self.layers: x = layer(x, mask) return self.norm(x) class EncoderLayer(nn.Module): """Encoder is made up of self-attn and feed forward (defined below)""" def __init__(self, size, self_attn, feed_forward, dropout): super(EncoderLayer, self).__init__() self.self_attn = self_attn self.feed_forward = feed_forward self.sublayer = clones(SublayerConnection(size, dropout), 2) self.size = size def forward(self, x, mask): """Follow Figure 1 (left) for connections.""" x = self.sublayer[0](x, lambda x: self.self_attn(x, x, x, mask)) return self.sublayer[1](x, self.feed_forward) ################ Decoder ################ class Decoder(nn.Module): """Generic N layer decoder with masking.""" def __init__(self, layer, N): super(Decoder, self).__init__() self.layers = clones(layer, N) self.norm = LayerNorm(layer.size) def forward(self, x, memory, src_mask, tgt_mask): for layer in self.layers: x = layer(x, memory, src_mask, tgt_mask) return self.norm(x) class DecoderLayer(nn.Module): """Decoder is made of self-attn, src-attn, and feed forward (defined below)""" def __init__(self, size, self_attn, src_attn, feed_forward, dropout): super(DecoderLayer, self).__init__() self.size = size self.self_attn = self_attn self.src_attn = src_attn self.feed_forward = feed_forward self.sublayer = clones(SublayerConnection(size, dropout), 3) def forward(self, x, memory, src_mask, tgt_mask): """Follow Figure 1 (right) for connections.""" m = memory x = self.sublayer[0](x, lambda x: self.self_attn(x, x, x, tgt_mask)) x = self.sublayer[1](x, lambda x: self.src_attn(x, m, m, src_mask)) return self.sublayer[2](x, self.feed_forward) ################ Generator ################ class Generator(nn.Module): """Define standard linear + softmax generation step.""" def __init__(self, d_model, vocab): super(Generator, self).__init__() self.proj = nn.Linear(d_model, vocab) def forward(self, x): return F.log_softmax(self.proj(x), dim=-1) class EncoderDecoder(nn.Module): """ A standard Encoder-Decoder architecture. Base for this and many other models. """ def __init__(self, encoder, decoder, src_embed, tgt_embed, generator, position_layer, model_size, latent_size): super(EncoderDecoder, self).__init__() self.encoder = encoder self.decoder = decoder self.src_embed = src_embed self.tgt_embed = tgt_embed self.generator = generator self.position_layer = position_layer self.model_size = model_size self.latent_size = latent_size self.sigmoid = nn.Sigmoid() # self.memory2latent = nn.Linear(self.model_size, self.latent_size) # self.latent2memory = nn.Linear(self.latent_size, self.model_size) def forward(self, src, tgt, src_mask, tgt_mask): """ Take in and process masked src and target sequences. """ latent = self.encode(src, src_mask) # (batch_size, max_src_seq, d_model) latent = self.sigmoid(latent) # memory = self.position_layer(memory) latent = torch.sum(latent, dim=1) # (batch_size, d_model) # latent = self.memory2latent(memory) # (batch_size, max_src_seq, latent_size) # latent = self.memory2latent(memory) # memory = self.latent2memory(latent) # (batch_size, max_src_seq, d_model) logit = self.decode(latent.unsqueeze(1), tgt, tgt_mask) # (batch_size, max_tgt_seq, d_model) prob = self.generator(logit) # (batch_size, max_seq, vocab_size) return latent, prob def encode(self, src, src_mask): return self.encoder(self.src_embed(src), src_mask) def decode(self, memory, tgt, tgt_mask): # memory: (batch_size, 1, d_model) src_mask = get_cuda(torch.ones(memory.size(0), 1, 1).long()) # print("src_mask here", src_mask) # print("src_mask", src_mask.size()) return self.decoder(self.tgt_embed(tgt), memory, src_mask, tgt_mask) def greedy_decode(self, latent, max_len, start_id): ''' latent: (batch_size, max_src_seq, d_model) src_mask: (batch_size, 1, max_src_len) ''' batch_size = latent.size(0) # memory = self.latent2memory(latent) ys = get_cuda(torch.ones(batch_size, 1).fill_(start_id).long()) # (batch_size, 1) for i in range(max_len - 1): # input("==========") # print("="*10, i) # print("ys", ys.size()) # (batch_size, i) # print("tgt_mask", subsequent_mask(ys.size(1)).size()) # (1, i, i) out = self.decode(latent.unsqueeze(1), to_var(ys), to_var(subsequent_mask(ys.size(1)).long())) prob = self.generator(out[:, -1]) # print("prob", prob.size()) # (batch_size, vocab_size) _, next_word = torch.max(prob, dim=1) # print("next_word", next_word.size()) # (batch_size) # print("next_word.unsqueeze(1)", next_word.unsqueeze(1).size()) ys = torch.cat([ys, next_word.unsqueeze(1)], dim=1) # print("ys", ys.size()) return ys[:, 1:] def make_model(d_vocab, N, d_model, latent_size, d_ff=1024, h=4, dropout=0.1): """Helper: Construct a model from hyperparameters.""" c = copy.deepcopy attn = MultiHeadedAttention(h, d_model) ff = PositionwiseFeedForward(d_model, d_ff, dropout) position = PositionalEncoding(d_model, dropout) share_embedding = Embeddings(d_model, d_vocab) model = EncoderDecoder( Encoder(EncoderLayer(d_model, c(attn), c(ff), dropout), N), Decoder(DecoderLayer(d_model, c(attn), c(attn), c(ff), dropout), N), # nn.Sequential(Embeddings(d_model, d_vocab), c(position)), # nn.Sequential(Embeddings(d_model, d_vocab), c(position)), nn.Sequential(share_embedding, c(position)), nn.Sequential(share_embedding, c(position)), Generator(d_model, d_vocab), c(position), d_model, latent_size, ) # This was important from their code. # Initialize parameters with Glorot / fan_avg. for p in model.parameters(): if p.dim() > 1: nn.init.xavier_uniform_(p) return model def subsequent_mask(size): "Mask out subsequent positions." attn_shape = (1, size, size) subsequent_mask = np.triu(np.ones(attn_shape), k=1).astype('uint8') return torch.from_numpy(subsequent_mask) == 0 class Batch: """Object for holding a batch of data with mask during training.""" def __init__(self, src, trg=None, pad=0): self.src = src self.src_mask = (src != pad).unsqueeze(-2) if trg is not None: self.trg = trg[:, :-1] self.trg_y = trg[:, 1:] self.trg_mask = self.make_std_mask(self.trg, pad) self.ntokens = (self.trg_y != pad).data.sum() @staticmethod def make_std_mask(tgt, pad): """Create a mask to hide padding and future words.""" tgt_mask = (tgt != pad).unsqueeze(-2) tgt_mask = tgt_mask & Variable( subsequent_mask(tgt.size(-1)).type_as(tgt_mask.data)) return tgt_mask class NoamOpt: "Optim wrapper that implements rate." def __init__(self, model_size, factor, warmup, optimizer): self.optimizer = optimizer self._step = 0 self.warmup = warmup self.factor = factor self.model_size = model_size self._rate = 0 def step(self): "Update parameters and rate" self._step += 1 rate = self.rate() for p in self.optimizer.param_groups: p['lr'] = rate self._rate = rate self.optimizer.step() def rate(self, step=None): "Implement `lrate` above" if step is None: step = self._step return self.factor * \ (self.model_size ** (-0.5) * min(step ** (-0.5), step * self.warmup ** (-1.5))) def get_std_opt(model): return NoamOpt(model.src_embed[0].d_model, 2, 4000, torch.optim.Adam(model.parameters(), lr=0, betas=(0.9, 0.98), eps=1e-9)) class LabelSmoothing(nn.Module): """Implement label smoothing.""" def __init__(self, size, padding_idx, smoothing=0.0): super(LabelSmoothing, self).__init__() self.criterion = nn.KLDivLoss(size_average=False) self.padding_idx = padding_idx self.confidence = 1.0 - smoothing self.smoothing = smoothing self.size = size self.true_dist = None def forward(self, x, target): assert x.size(1) == self.size true_dist = x.data.clone() true_dist.fill_(self.smoothing / (self.size - 2)) true_dist.scatter_(1, target.data.unsqueeze(1), self.confidence) true_dist[:, self.padding_idx] = 0 mask = torch.nonzero(target.data == self.padding_idx) if mask.dim() > 0: true_dist.index_fill_(0, mask.squeeze(), 0.0) self.true_dist = true_dist return self.criterion(x, Variable(true_dist, requires_grad=False)) class Classifier(nn.Module): def __init__(self, latent_size, output_size): super().__init__() self.fc1 = nn.Linear(latent_size, 100) self.relu1 = nn.LeakyReLU(0.2, ) self.fc2 = nn.Linear(100, 50) self.relu2 = nn.LeakyReLU(0.2) self.fc3 = nn.Linear(50, output_size) self.sigmoid = nn.Sigmoid() def forward(self, input): out = self.fc1(input) out = self.relu1(out) out = self.fc2(out) out = self.relu2(out) out = self.fc3(out) out = self.sigmoid(out) # out = F.log_softmax(out, dim=1) return out # batch_size * label_size def fgim_attack(model, origin_data, target, ae_model, max_sequence_length, id_bos, id2text_sentence, id_to_word, gold_ans): """Fast Gradient Iterative Methods""" dis_criterion = nn.BCELoss(size_average=True) gold_text = id2text_sentence(gold_ans, id_to_word) print("gold:", gold_text) # while True: for epsilon in [2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0]: it = 0 data = origin_data while True: print("epsilon:", epsilon) data = to_var(data.clone()) # (batch_size, seq_length, latent_size) # Set requires_grad attribute of tensor. Important for Attack data.requires_grad = True output = model.forward(data) # Calculate gradients of model in backward pass # print("target", target[0].item()) # print("output", output[0].item()) loss = dis_criterion(output, target) model.zero_grad() # dis_optimizer.zero_grad() loss.backward() data_grad = data.grad.data # print("data_grad") # print(data_grad) data = data - epsilon * data_grad # print("epsilon * data_grad") # print((epsilon * data_grad)) # print("data") # print(data) # print("perturbed_data") # print(perturbed_data) it += 1 # data = perturbed_data epsilon = epsilon * 0.9 generator_id = ae_model.greedy_decode(data, max_len=max_sequence_length, start_id=id_bos) generator_text = id2text_sentence(generator_id[0], id_to_word) print("| It {:2d} | dis model pred {:5.4f} |".format(it, output[0].item())) print(generator_text) if it >= 5: break return if __name__ == '__main__': # plt.figure(figsize=(15, 5)) # pe = PositionalEncoding(20, 0) # y = pe.forward(Variable(torch.zeros(1, 100, 20))) # plt.plot(np.arange(100), y[0, :, 4:8].data.numpy()) # plt.legend(["dim %d" % p for p in [4, 5, 6, 7]]) # plt.show() # Small example model. # tmp_model = make_model(10, 10, 2) pass