--- license: apache-2.0 --- **Transformer_translation** ``` import torch import warnings warnings.filterwarnings("ignore") import torch import torch.nn as nn import math from transformers import MarianTokenizer from datasets import load_dataset from typing import List from torch import Tensor from torch.nn import Transformer from torch.nn.utils.rnn import pad_sequence from torch.utils.data import DataLoader, Dataset from timeit import default_timer as timer import urllib.request import os from torch.cuda.amp import GradScaler, autocast import logging tokenizer = MarianTokenizer.from_pretrained('Helsinki-NLP/opus-mt-de-en') # 定义特殊token的索引 PAD_IDX = tokenizer.pad_token_id BOS_IDX = tokenizer.bos_token_id if tokenizer.bos_token_id is not None else 1 EOS_IDX = tokenizer.eos_token_id UNK_IDX = tokenizer.unk_token_id class PositionalEncoding(nn.Module): def __init__(self, emb_size: int, dropout: float, maxlen: int = 5000): super(PositionalEncoding, self).__init__() den = torch.exp(-torch.arange(0, emb_size, 2) * math.log(10000) / emb_size) pos = torch.arange(0, maxlen).reshape(maxlen, 1) pos_embedding = torch.zeros((maxlen, emb_size)) pos_embedding[:, 0::2] = torch.sin(pos * den) pos_embedding[:, 1::2] = torch.cos(pos * den) pos_embedding = pos_embedding.unsqueeze(-2) self.dropout = nn.Dropout(dropout) self.register_buffer('pos_embedding', pos_embedding) def forward(self, token_embedding: Tensor): return self.dropout(token_embedding + self.pos_embedding[:token_embedding.size(0), :]) class TokenEmbedding(nn.Module): def __init__(self, vocab_size: int, emb_size): super(TokenEmbedding, self).__init__() self.embedding = nn.Embedding(vocab_size, emb_size) self.emb_size = emb_size def forward(self, tokens: Tensor): return self.embedding(tokens.long()) * math.sqrt(self.emb_size) class Seq2SeqTransformer(nn.Module): def __init__(self, num_encoder_layers: int, num_decoder_layers: int, emb_size: int, nhead: int, src_vocab_size: int, tgt_vocab_size: int, dim_feedforward: int = 512, dropout: float = 0.1): super(Seq2SeqTransformer, self).__init__() self.transformer = Transformer(d_model=emb_size, nhead=nhead, num_encoder_layers=num_encoder_layers, num_decoder_layers=num_decoder_layers, dim_feedforward=dim_feedforward, dropout=dropout) self.generator = nn.Linear(emb_size, tgt_vocab_size) self.src_tok_emb = TokenEmbedding(src_vocab_size, emb_size) self.tgt_tok_emb = TokenEmbedding(tgt_vocab_size, emb_size) self.positional_encoding = PositionalEncoding(emb_size, dropout=dropout) def forward(self, src: Tensor, trg: Tensor, src_mask: Tensor, tgt_mask: Tensor, src_padding_mask: Tensor, tgt_padding_mask: Tensor, memory_key_padding_mask: Tensor): src_emb = self.positional_encoding(self.src_tok_emb(src)) tgt_emb = self.positional_encoding(self.tgt_tok_emb(trg)) outs = self.transformer(src_emb, tgt_emb, src_mask, tgt_mask, None, src_padding_mask, tgt_padding_mask, memory_key_padding_mask) return self.generator(outs) def encode(self, src: Tensor, src_mask: Tensor): return self.transformer.encoder(self.positional_encoding(self.src_tok_emb(src)), src_mask) def decode(self, tgt: Tensor, memory: Tensor, tgt_mask: Tensor): return self.transformer.decoder(self.positional_encoding(self.tgt_tok_emb(tgt)), memory, tgt_mask) def generate_square_subsequent_mask(sz): mask = (torch.triu(torch.ones((sz, sz), device=DEVICE)) == 1).transpose(0, 1) mask = mask.float().masked_fill(mask == 0, float('-inf')).masked_fill(mask == 1, float(0.0)) return mask def create_mask(src, tgt): src_seq_len = src.shape[0] tgt_seq_len = tgt.shape[0] tgt_mask = generate_square_subsequent_mask(tgt_seq_len) src_mask = torch.zeros((src_seq_len, src_seq_len), device=DEVICE).type(torch.bool) src_padding_mask = (src == PAD_IDX).transpose(0, 1) tgt_padding_mask = (tgt == PAD_IDX).transpose(0, 1) return src_mask, tgt_mask, src_padding_mask, tgt_padding_mask def greedy_decode(model, src, src_mask, max_len, start_symbol): src = src.to(DEVICE) src_mask = src_mask.to(DEVICE) memory = model.encode(src, src_mask) ys = torch.ones(1, 1).fill_(start_symbol).type(torch.long).to(DEVICE) for i in range(max_len-1): memory = memory.to(DEVICE) tgt_mask = (generate_square_subsequent_mask(ys.size(0)) .type(torch.bool)).to(DEVICE) out = model.decode(ys, memory, tgt_mask) out = out.transpose(0, 1) prob = model.generator(out[:, -1]) _, next_word = torch.max(prob, dim=1) next_word = next_word.item() ys = torch.cat([ys, torch.ones(1, 1).type_as(src.data).fill_(next_word)], dim=0) if next_word == EOS_IDX: break return ys def translate(model: torch.nn.Module, src_sentence: str): model.eval() tokens = tokenizer(src_sentence, return_tensors='pt', padding=True) src = tokens['input_ids'].transpose(0, 1).to(DEVICE) src_mask = (torch.zeros(src.shape[0], src.shape[0])).type(torch.bool).to(DEVICE) tgt_tokens = greedy_decode(model, src, src_mask, max_len=src.shape[0] + 5, start_symbol=BOS_IDX).flatten() return tokenizer.decode(tgt_tokens.tolist(), skip_special_tokens=True) BATCH_SIZE = 32 # 减小批次大小，原来是64 EMB_SIZE = 512 # 保持不变 NHEAD = 8 # 保持不变 FFN_HID_DIM = 512 # 改回512，原来改成了1024 NUM_ENCODER_LAYERS = 3 # 改回3，原来改成了4 NUM_DECODER_LAYERS = 3 # 改回3，原来改成了4 NUM_EPOCHS = 18 # 保持不变 SRC_VOCAB_SIZE = tokenizer.vocab_size TGT_VOCAB_SIZE = tokenizer.vocab_size DEVICE = 'cuda' transformer = Seq2SeqTransformer(NUM_ENCODER_LAYERS, NUM_DECODER_LAYERS, EMB_SIZE, NHEAD, SRC_VOCAB_SIZE, TGT_VOCAB_SIZE, FFN_HID_DIM) transformer.load_state_dict(torch.load('./transformer_translation.pth')) transformer = transformer.to(DEVICE) print('开始翻译...') print(translate(transformer, "Eine Gruppe von Freunden spielt Billiade.")) print('翻译完成！！') ```