用注意力机制实现中英文互译
用注意力机制实现中英文互译
[KEY: > input, = target, < output]
il est en train de peindre un tableau .
= he is painting a picture .
< he is painting a picture .
pourquoi ne pas essayer ce vin delicieux ?
= why not try that delicious wine ?
< why not try that delicious wine ?
elle n est pas poete mais romanciere .
= she is not a poet but a novelist .
< she not not a poet but a novelist .
导入需要的模块及数据
from __future__ import unicode_literals, print_function, division
from io import open
import unicodedata
import string
import re
import random
import jieba
import torch
import torch.nn as nn
from torch import optim
import torch.nn.functional as Fimport matplotlib.font_manager as fm
myfont = fm.FontProperties(fname='/Users/maqi/opt/anaconda3/lib/python3.8/site-packages/matplotlib/mpl-data/fonts/ttf/DejaVuSans.ttf')device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
预处理数据
SOS_token = 0
EOS_token = 1class Lang:def __init__(self, name):self.name = nameself.word2index = {}self.word2count = {}self.index2word = {0: "SOS", 1: "EOS"}self.n_words = 2 # Count SOS and EOSdef addSentence(self, sentence):for word in sentence.split(' '):self.addWord(word)def addSentence_cn(self, sentence):for word in list(jieba.cut(sentence)):self.addWord(word)def addWord(self, word):if word not in self.word2index:self.word2index[word] = self.n_wordsself.word2count[word] = 1self.index2word[self.n_words] = wordself.n_words += 1else:self.word2count[word] += 1
# 为便于数据处理,把Unicode字符串转换为ASCII编码def unicodeToAscii(s):return ''.join(c for c in unicodedata.normalize('NFD', s)if unicodedata.category(c) != 'Mn')# 对英文转换为小写,去空格及非字母符号等处理def normalizeString(s):s = unicodeToAscii(s.lower().strip())s = re.sub(r"([.!?])", r" \1", s)#s = re.sub(r"[^a-zA-Z.!?]+", r" ", s)return s
def readLangs(lang1, lang2, reverse=False):print("Reading lines...")# 读文件,然后分成行lines = open('eng-cmn/%s-%s.txt' % (lang1, lang2), encoding='utf-8').\read().strip().split('\n')# 把行分成语句对,并进行规范化pairs = [[normalizeString(s) for s in l.split('\t')] for l in lines]# 判断是否需要转换语句对的次序,如[英文,中文]转换为[中文,英文]次序if reverse:pairs = [list(reversed(p)) for p in pairs]input_lang = Lang(lang2)output_lang = Lang(lang1)else:input_lang = Lang(lang1)output_lang = Lang(lang2)return input_lang, output_lang, pairs
#为便于训练,这里选择部分数据
MAX_LENGTH = 20eng_prefixes = ("i am ", "i m ","he is", "he s ","she is", "she s ","you are", "you re ","we are", "we re ","they are", "they re "
)def filterPair(p):return len(p[0].split(' ')) < MAX_LENGTH and \len(p[1].split(' ')) < MAX_LENGTH and \p[1].startswith(eng_prefixes)def filterPairs(pairs):return [pair for pair in pairs if filterPair(pair)]
def prepareData(lang1, lang2, reverse=False):input_lang, output_lang, pairs = readLangs(lang1, lang2, reverse)print("Read %s sentence pairs" % len(pairs))pairs = filterPairs(pairs)print("Trimmed to %s sentence pairs" % len(pairs))print("Counting words...")for pair in pairs:input_lang.addSentence_cn(pair[0])output_lang.addSentence(pair[1])print("Counted words:")print(input_lang.name, input_lang.n_words)print(output_lang.name, output_lang.n_words)return input_lang, output_lang, pairs
input_lang, output_lang, pairs = prepareData('eng', 'cmn',True)
print(random.choice(pairs))
Reading lines...Building prefix dict from the default dictionary ...Loading model from cache /var/folders/7t/wvjcfn5575g892qb2nqbd9kw0000gn/T/jieba.cacheRead 21007 sentence pairsTrimmed to 640 sentence pairsCounting words...Loading model cost 0.571 seconds.Prefix dict has been built succesfully.Counted words:cmn 1063eng 808['他很穷。', 'he is poor .']
pairs[:3]
[['我冷。', 'i am cold .'], ['我沒事。', 'i am okay .'], ['我生病了。', 'i am sick .']]
构建模型
class EncoderRNN(nn.Module): def __init__(self, input_size, hidden_size): super(EncoderRNN, self).__init__() self.hidden_size = hidden_size self.embedding = nn.Embedding(input_size, hidden_size) self.gru = nn.GRU(hidden_size, hidden_size) def forward(self, input, hidden): embedded = self.embedding(input).view(1, 1, -1) output = embedded output, hidden = self.gru(output, hidden) return output, hidden def initHidden(self): return torch.zeros(1, 1, self.hidden_size, device=device)
class DecoderRNN(nn.Module): def __init__(self, hidden_size, output_size): super(DecoderRNN, self).__init__() self.hidden_size = hidden_size self.embedding = nn.Embedding(output_size, hidden_size) self.gru = nn.GRU(hidden_size, hidden_size) self.out = nn.Linear(hidden_size, output_size) self.softmax = nn.LogSoftmax(dim=1) def forward(self, input, hidden): output = self.embedding(input).view(1, 1, -1) output = F.relu(output) output, hidden = self.gru(output, hidden) output = self.softmax(self.out(output[0])) return output, hidden def initHidden(self): return torch.zeros(1, 1, self.hidden_size, device=device)
class AttnDecoderRNN(nn.Module): def __init__(self, hidden_size, output_size, dropout_p=0.1, max_length=MAX_LENGTH): super(AttnDecoderRNN, self).__init__() self.hidden_size = hidden_size self.output_size = output_size self.dropout_p = dropout_p self.max_length = max_length self.embedding = nn.Embedding(self.output_size, self.hidden_size) self.attn = nn.Linear(self.hidden_size * 2, self.max_length) self.attn_combine = nn.Linear(self.hidden_size * 2, self.hidden_size) self.dropout = nn.Dropout(self.dropout_p) self.gru = nn.GRU(self.hidden_size, self.hidden_size) self.out = nn.Linear(self.hidden_size, self.output_size) def forward(self, input, hidden, encoder_outputs): embedded = self.embedding(input).view(1, 1, -1) embedded = self.dropout(embedded) attn_weights = F.softmax( self.attn(torch.cat((embedded[0], hidden[0]), 1)), dim=1) attn_applied = torch.bmm(attn_weights.unsqueeze(0), encoder_outputs.unsqueeze(0)) output = torch.cat((embedded[0], attn_applied[0]), 1) output = self.attn_combine(output).unsqueeze(0) output = F.relu(output) output, hidden = self.gru(output, hidden) output = F.log_softmax(self.out(output[0]), dim=1) return output, hidden, attn_weights def initHidden(self): return torch.zeros(1, 1, self.hidden_size, device=device)
def indexesFromSentence(lang, sentence): return [lang.word2index[word] for word in sentence.split(' ')]def indexesFromSentence_cn(lang, sentence): return [lang.word2index[word] for word in list(jieba.cut(sentence))]def tensorFromSentence(lang, sentence): indexes = indexesFromSentence(lang, sentence) indexes.append(EOS_token) return torch.tensor(indexes, dtype=torch.long, device=device).view(-1, 1)def tensorFromSentence_cn(lang, sentence): indexes = indexesFromSentence_cn(lang, sentence) indexes.append(EOS_token) return torch.tensor(indexes, dtype=torch.long, device=device).view(-1, 1)def tensorsFromPair(pair): input_tensor = tensorFromSentence_cn(input_lang, pair[0]) target_tensor = tensorFromSentence(output_lang, pair[1]) return (input_tensor, target_tensor)
训练模型
teacher_forcing_ratio = 0.5def train(input_tensor, target_tensor, encoder, decoder, encoder_optimizer, decoder_optimizer, criterion, max_length=MAX_LENGTH): encoder_hidden = encoder.initHidden() encoder_optimizer.zero_grad() decoder_optimizer.zero_grad() input_length = input_tensor.size(0) target_length = target_tensor.size(0) encoder_outputs = torch.zeros(max_length, encoder.hidden_size, device=device) loss = 0 for ei in range(input_length): encoder_output, encoder_hidden = encoder( input_tensor[ei], encoder_hidden) encoder_outputs[ei] = encoder_output[0, 0] decoder_input = torch.tensor([[SOS_token]], device=device) decoder_hidden = encoder_hidden use_teacher_forcing = True if random.random() < teacher_forcing_ratio else False if use_teacher_forcing: # Teacher forcing: Feed the target as the next input for di in range(target_length): decoder_output, decoder_hidden, decoder_attention = decoder( decoder_input, decoder_hidden, encoder_outputs) loss += criterion(decoder_output, target_tensor[di]) decoder_input = target_tensor[di] # Teacher forcing else: # Without teacher forcing: use its own predictions as the next input for di in range(target_length): decoder_output, decoder_hidden, decoder_attention = decoder( decoder_input, decoder_hidden, encoder_outputs) topv, topi = decoder_output.topk(1) decoder_input = topi.squeeze().detach() # detach from history as input loss += criterion(decoder_output, target_tensor[di]) if decoder_input.item() == EOS_token: break loss.backward() encoder_optimizer.step() decoder_optimizer.step() return loss.item() / target_length
import timeimport mathdef asMinutes(s): m = math.floor(s / 60) s -= m * 60 return '%dm %ds' % (m, s)def timeSince(since, percent): now = time.time() s = now - since es = s / (percent) rs = es - s return '%s (- %s)' % (asMinutes(s), asMinutes(rs))
def trainIters(encoder, decoder, n_iters, print_every=1000, plot_every=100, learning_rate=0.01): start = time.time() plot_losses = [] print_loss_total = 0 plot_loss_total = 0 encoder_optimizer = optim.SGD(encoder.parameters(), lr=learning_rate) decoder_optimizer = optim.SGD(decoder.parameters(), lr=learning_rate) training_pairs = [tensorsFromPair(random.choice(pairs)) for i in range(n_iters)] criterion = nn.NLLLoss() for iter in range(1, n_iters + 1): training_pair = training_pairs[iter - 1] input_tensor = training_pair[0] target_tensor = training_pair[1] loss = train(input_tensor, target_tensor, encoder, decoder, encoder_optimizer, decoder_optimizer, criterion) print_loss_total += loss plot_loss_total += loss if iter % print_every == 0: print_loss_avg = print_loss_total / print_every print_loss_total = 0 print('%s (%d %d%%) %.4f' % (timeSince(start, iter / n_iters), iter, iter / n_iters * 100, print_loss_avg)) if iter % plot_every == 0: plot_loss_avg = plot_loss_total / plot_every plot_losses.append(plot_loss_avg) plot_loss_total = 0 showPlot(plot_losses)
import matplotlib.pyplot as plt%matplotlib inline#plt.switch_backend('agg')import matplotlib.ticker as tickerimport numpy as npdef showPlot(points): plt.figure() fig, ax = plt.subplots() # this locator puts ticks at regular intervals loc = ticker.MultipleLocator(base=0.2) ax.yaxis.set_major_locator(loc) plt.plot(points)
def evaluate(encoder, decoder, sentence, max_length=MAX_LENGTH): with torch.no_grad(): input_tensor = tensorFromSentence_cn(input_lang, sentence) input_length = input_tensor.size()[0] encoder_hidden = encoder.initHidden() encoder_outputs = torch.zeros(max_length, encoder.hidden_size, device=device) for ei in range(input_length): encoder_output, encoder_hidden = encoder(input_tensor[ei], encoder_hidden) encoder_outputs[ei] += encoder_output[0, 0] decoder_input = torch.tensor([[SOS_token]], device=device) # SOS decoder_hidden = encoder_hidden decoded_words = [] decoder_attentions = torch.zeros(max_length, max_length) for di in range(max_length): decoder_output, decoder_hidden, decoder_attention = decoder( decoder_input, decoder_hidden, encoder_outputs) decoder_attentions[di] = decoder_attention.data topv, topi = decoder_output.data.topk(1) if topi.item() == EOS_token: decoded_words.append('<EOS>') break else: decoded_words.append(output_lang.index2word[topi.item()]) decoder_input = topi.squeeze().detach() return decoded_words, decoder_attentions[:di + 1]
def evaluateRandomly(encoder, decoder, n=10): for i in range(n): pair = random.choice(pairs) print('>', pair[0]) print('=', pair[1]) output_words, attentions = evaluate(encoder, decoder, pair[0]) output_sentence = ' '.join(output_words) print('<', output_sentence) print('')
hidden_size = 256encoder1 = EncoderRNN(input_lang.n_words, hidden_size).to(device)attn_decoder1 = AttnDecoderRNN(hidden_size, output_lang.n_words, dropout_p=0.1).to(device)trainIters(encoder1, attn_decoder1, 75000, print_every=5000)
1m 54s (- 26m 36s) (5000 6%) 2.63943m 43s (- 24m 10s) (10000 13%) 1.09165m 34s (- 22m 19s) (15000 20%) 0.20577m 29s (- 20m 36s) (20000 26%) 0.04459m 27s (- 18m 54s) (25000 33%) 0.025311m 25s (- 17m 7s) (30000 40%) 0.020213m 20s (- 15m 14s) (35000 46%) 0.017515m 17s (- 13m 23s) (40000 53%) 0.016717m 15s (- 11m 30s) (45000 60%) 0.014119m 13s (- 9m 36s) (50000 66%) 0.013721m 12s (- 7m 42s) (55000 73%) 0.011023m 12s (- 5m 48s) (60000 80%) 0.011625m 12s (- 3m 52s) (65000 86%) 0.012527m 11s (- 1m 56s) (70000 93%) 0.009129m 11s (- 0m 0s) (75000 100%) 0.0095<Figure size 432x288 with 0 Axes>
随机采样,对模型进行测试
evaluateRandomly(encoder1, attn_decoder1)
> 今天下午我會外出。
= i am going out this afternoon .
< i am going out this afternoon . <EOS>> 我相信他是無辜的。
= i am convinced that he is innocent .
< i am convinced that he is innocent . <EOS>> 他在自己房里玩。
= he is playing in his room .
< he is playing in his room . <EOS>> 我來自四國。
= i am from shikoku .
< i am from shikoku . <EOS>> 她戴著一頂帽子。
= she is wearing a hat .
< she is wearing a hat . <EOS>> 您非常勇敢。
= you are very courageous .
< you are very brave . <EOS>> 他有几分像学者。
= he is something of a scholar .
< he is something of a scholar . <EOS>> 你真傻。
= you are so stupid .
< you are so stupid . <EOS>> 他年紀夠大可以瞭解它。
= he is old enough to understand it .
< he is old enough to understand it . <EOS>> 你別小看了他。
= you are selling him short .
< you are selling him short . <EOS>
def evaluate_randomly():pair = random.choice(pairs)output_words, decoder_attn = evaluate(pair[0])output_sentence = ' '.join(output_words)print('>', pair[0])print('=', pair[1])print('<', output_sentence)print('')
def evaluateRandomly(encoder, decoder, n=20):for i in range(n):pair = random.choice(pairs)print('>', pair[0])print('=', pair[1])output_words, attentions = evaluate(encoder, decoder, pair[0])output_sentence = ' '.join(output_words)print('<', output_sentence)print('')
可视化注意力
def showAttention(input_sentence, output_words, attentions):# Set up figure with colorbarfig = plt.figure()ax = fig.add_subplot(111)cax = ax.matshow(attentions.numpy(), cmap='bone')fig.colorbar(cax)# Set up axesax.set_xticklabels([''] + list(jieba.cut(input_sentence)) +['<EOS>'], rotation=90,fontproperties=myfont)ax.set_yticklabels([''] + output_words)# Show label at every tickax.xaxis.set_major_locator(ticker.MultipleLocator(1))ax.yaxis.set_major_locator(ticker.MultipleLocator(1))plt.show()
def evaluateAndShowAttention(input_sentence):output_words, attentions = evaluate(encoder1, attn_decoder1, input_sentence)print('input =', input_sentence)print('output =', ' '.join(output_words))showAttention(input_sentence, output_words, attentions)evaluateAndShowAttention("我很幸福。")evaluateAndShowAttention("我们在严肃地谈论你的未来。")evaluateAndShowAttention("我在家。")evaluateAndShowAttention("我们在严肃地谈论你的未来。")
input = 我很幸福。
output = i am very happy . <EOS><ipython-input-23-2d6791f485ef>:9: UserWarning: FixedFormatter should only be used together with FixedLocatorax.set_xticklabels([''] + list(jieba.cut(input_sentence)) +
<ipython-input-23-2d6791f485ef>:11: UserWarning: FixedFormatter should only be used together with FixedLocatorax.set_yticklabels([''] + output_words)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 25105 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 24456 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 24184 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 31119 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 12290 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 25105 missing from current font.font.set_text(s, 0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 24456 missing from current font.font.set_text(s, 0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 24184 missing from current font.font.set_text(s, 0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 31119 missing from current font.font.set_text(s, 0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 12290 missing from current font.font.set_text(s, 0, flags=flags)
input = 我们在严肃地谈论你的未来。
output = we are having a serious talk about your future . <EOS>/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 20204 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 22312 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 20005 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 32899 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 22320 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 35848 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 35770 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 20320 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 30340 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 26410 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 26469 missing from current font.font.set_text(s, 0.0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 20204 missing from current font.font.set_text(s, 0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 22312 missing from current font.font.set_text(s, 0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 20005 missing from current font.font.set_text(s, 0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 32899 missing from current font.font.set_text(s, 0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 22320 missing from current font.font.set_text(s, 0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 35848 missing from current font.font.set_text(s, 0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 35770 missing from current font.font.set_text(s, 0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 20320 missing from current font.font.set_text(s, 0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 30340 missing from current font.font.set_text(s, 0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 26410 missing from current font.font.set_text(s, 0, flags=flags)
/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 26469 missing from current font.font.set_text(s, 0, flags=flags)
input = 我在家。output = i am at home . <EOS>/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:240: RuntimeWarning: Glyph 23478 missing from current font. font.set_text(s, 0.0, flags=flags)/Users/maqi/opt/anaconda3/envs/mq_env/lib/python3.8/site-packages/matplotlib/backends/backend_agg.py:203: RuntimeWarning: Glyph 23478 missing from current font. font.set_text(s, 0, flags=flags)
input = 我们在严肃地谈论你的未来。output = we are having a serious talk about your future . <EOS>
用注意力机制实现中英文互译相关推荐
- 机器翻译:引入注意力机制的Encoder-Decoder深度神经网络训练实战中英文互译(完结篇)
文章目录 前言 一.数据的预处理 二.模型配置 三.模型训练 总结 前言 注意力机制是一种在深度学习中广泛使用的技术,它可以有效地处理输入序列和输出序列之间的对应关系.其中,Encoder-Decod ...
- 中英文互译在线翻译-在线翻译软件
中英文互译在线翻译哪个软件好用,今天给大家分享一款免费好用的批量翻译器,汇集了世界上最好的几个翻译平台(百度/谷歌/有道),为什么这么多人使用它?首先第一点翻译质量高,选择性多.第二点支持各种语言互译 ...
- Learning Hammerspoon中英文互译
Learning Hammerspoon中英文互译 本书主要介绍mac的脚本软件Hammerspoon的基础使用,仅提供可阅读部分翻译. Unleash the power of automation ...
- 计算机英语短文互译,中英文互译的英语短文
在英语学习中,阅读能力是学习者发展其它语言能力(听.说.写.译)的基础.阅读能力的高低,不仅决定了学习者获取知识和信息的水平,而且在一定程度上也反映出学习者综合运用英语的能力.小编精心收集了中英文互译 ...
- 多语种翻译器,中英文互译翻译器
中英文互译翻译器可以批量处理文档,在线本地文档中英文互译翻译,多格式多语言输入本地批量导出.通过谷歌翻译.百度翻译和有道词典等api接口对接,中文.英文.日文.意大利语等多个语种可以互译. 中英文互译 ...
- 中英文互译之Excel表格
在做英文数据库时,进行数据去重整理会在excel表格中操作. 很多英文标题或摘要看不懂,需要复制到翻译软件中进行翻译. 有没有办法可以直接在表格中给出翻译呢? 基于此,COOC开发了这一功能,提取出的 ...
- 计算机网络中英互译,计算机网络中英文互译.doc
--- - 计算机网络中英翻译 ACK (ACKnowledgement) 确 认 帧 ADSL (Asymmetric Digital Subscriber Line) 非 对 称 数 字 用 户 ...
- 二极管名称中英文互译
abrupt junction diode 突变结二极管 absorber diode 吸收二极管 alkali ion diode 碱离子二极管 alloy junction diode 合金结二极 ...
- cad怎么把图层英文变成中文_CAD图层标准中英文互译
The CAD Layer Names DWS 标准文件简写专业码 English Name 英文名 Chinese Name 中 文名 Description 描述 Abbreviate 缩写 ...
最新文章
- Flutter控件--Switch 和 SwitchListTile
- MPB:南土所褚海燕组-​非靶标代谢组测定土壤可提取有机碳组分
- 计算机主键盘的布局,计算机键盘的布局,结构和布置
- Python线程同步机制: Locks, RLocks, Semaphores, Condition
- STM32之option bytes踩坑记录
- 【数据结构作业—01】用单循环链表解决约瑟夫问题
- Redis的两种备份方式:RDB和AOF
- C++获取文本文件字节数的一个小方法
- 数据库悲观锁和乐观锁
- 神经网络加速器设计研究:寒武纪DaDianNao论文阅读
- 程序员也要学英语——动词相关合集
- vue项目中通过cdn引入资源并配置
- 本机号码一键登录原理与应用(荣耀典藏版)
- 互信息(Mutual Information)的介绍
- 夜游灯光秀如何激活城市经济发展
- 用阿里云搭建Http代理服务器
- C++ 解决rand()函数生成的随机数每次都一样的问题
- BrainAGE作为大脑老化的神经影像标志物的十年
- 《游戏的人》笔记——第一章
- unturned服务器怎么自定义,《Unturned》机房服务器开服方法图文教程