pytorch bilstm crf NER

2023-12-18 00:08
文章标签 pytorch bilstm crf ner

本文主要是介绍pytorch bilstm crf NER，希望对大家解决编程问题提供一定的参考价值，需要的开发者们随着小编来一起学习吧！

参考链接：https://blog.csdn.net/Jason__Liang/article/details/81772632

https://github.com/jayavardhanr/End-to-end-Sequence-Labeling-via-Bi-directional-LSTM-CNNs-CRF-Tutorial/blob/master/Named_Entity_Recognition-LSTM-CNN-CRF-Tutorial.ipynb

 


import torch
import torch.autograd as autograd
import torch.nn as nn
import torch.optim as optimtorch.manual_seed(1)def argmax(vec):# return the argmax as a python int_, idx = torch.max(vec, 1)return idx.item()def prepare_sequence(seq, to_ix):idxs = [to_ix[w] for w in seq]return torch.tensor(idxs, dtype=torch.long)# Compute log sum exp in a numerically stable way for the forward algorithm
def log_sum_exp(vec):max_score = vec[0, argmax(vec)]max_score_broadcast = max_score.view(1, -1).expand(1, vec.size()[1])return max_score + \torch.log(torch.sum(torch.exp(vec - max_score_broadcast)))#####################################################################
# Create modelclass BiLSTM_CRF(nn.Module):def __init__(self, vocab_size, tag_to_ix, embedding_dim, hidden_dim):super(BiLSTM_CRF, self).__init__()self.embedding_dim = embedding_dimself.hidden_dim = hidden_dimself.vocab_size = vocab_sizeself.tag_to_ix = tag_to_ixself.tagset_size = len(tag_to_ix)self.word_embeds = nn.Embedding(vocab_size, embedding_dim)self.lstm = nn.LSTM(embedding_dim, hidden_dim // 2,num_layers=1, bidirectional=True)# Maps the output of the LSTM into tag space.self.hidden2tag = nn.Linear(hidden_dim, self.tagset_size)# Matrix of transition parameters.  Entry i,j is the score of# transitioning *to* i *from* j.self.transitions = nn.Parameter(torch.randn(self.tagset_size, self.tagset_size))# These two statements enforce the constraint that we never transfer# to the start tag and we never transfer from the stop tagself.transitions.data[tag_to_ix[START_TAG], :] = -10000self.transitions.data[:, tag_to_ix[STOP_TAG]] = -10000self.hidden = self.init_hidden()def init_hidden(self):return (torch.randn(2, 1, self.hidden_dim // 2),torch.randn(2, 1, self.hidden_dim // 2))def _forward_alg(self, feats):# Do the forward algorithm to compute the partition functioninit_alphas = torch.full((1, self.tagset_size), -10000.)# START_TAG has all of the score.init_alphas[0][self.tag_to_ix[START_TAG]] = 0.# Wrap in a variable so that we will get automatic backpropforward_var = init_alphas# Iterate through the sentencefor feat in feats:alphas_t = []  # The forward tensors at this timestepfor next_tag in range(self.tagset_size):# broadcast the emission score: it is the same regardless of# the previous tagemit_score = feat[next_tag].view(1, -1).expand(1, self.tagset_size)# the ith entry of trans_score is the score of transitioning to# next_tag from itrans_score = self.transitions[next_tag].view(1, -1)# The ith entry of next_tag_var is the value for the# edge (i -> next_tag) before we do log-sum-expnext_tag_var = forward_var + trans_score + emit_score# The forward variable for this tag is log-sum-exp of all the# scores.alphas_t.append(log_sum_exp(next_tag_var).view(1))forward_var = torch.cat(alphas_t).view(1, -1)terminal_var = forward_var + self.transitions[self.tag_to_ix[STOP_TAG]]alpha = log_sum_exp(terminal_var)return alphadef _get_lstm_features(self, sentence):# 随机初始化 num_layers*num_directions = 2, batch = 1, hidden_size = self.hidden_dim // 2self.hidden = self.init_hidden()# 为了满足input输入格式  embedding_dim 是 5embeds = self.word_embeds(sentence).view(len(sentence), 1, -1)# 输入 embeds -> 【11， 1， 5】、 self.hidden -> 是一个tuple ([], [])"""输入数据格式：input(seq_len, batch, input_size)h0(num_layers * num_directions, batch, hidden_size)c0(num_layers * num_directions, batch, hidden_size)输出数据格式：output(seq_len, batch, hidden_size * num_directions)hn(num_layers * num_directions, batch, hidden_size)cn(num_layers * num_directions, batch, hidden_size)"""lstm_out, self.hidden = self.lstm(embeds, self.hidden)lstm_out = lstm_out.view(len(sentence), self.hidden_dim)  # 回到原来的格式, 去掉batch_size# 从 [11, 4] -> [11, 5] # 最后的类别是5个lstm_feats = self.hidden2tag(lstm_out)return lstm_featsdef _score_sentence(self, feats, tags):# Gives the score of a provided tag sequencescore = torch.zeros(1)tags = torch.cat([torch.tensor([self.tag_to_ix[START_TAG]], dtype=torch.long), tags])for i, feat in enumerate(feats):score = score + \self.transitions[tags[i + 1], tags[i]] + feat[tags[i + 1]]score = score + self.transitions[self.tag_to_ix[STOP_TAG], tags[-1]]return scoredef _viterbi_decode(self, feats):backpointers = []# Initialize the viterbi variables in log spaceinit_vvars = torch.full((1, self.tagset_size), -10000.)init_vvars[0][self.tag_to_ix[START_TAG]] = 0# forward_var at step i holds the viterbi variables for step i-1forward_var = init_vvarsfor feat in feats:  # 对每个word进行循环, feat是每个word的emission分数bptrs_t = []  # holds the backpointers for this stepviterbivars_t = []  # holds the viterbi variables for this stepfor next_tag in range(self.tagset_size):  # 对每个tag进行循环# next_tag_var[i] holds the viterbi variable for tag i at the# previous step, plus the score of transitioning# from tag i to next_tag.# We don't include the emission scores here because the max# does not depend on them (we add them in below)next_tag_var = forward_var + self.transitions[next_tag]  # 这里 加了转移矩阵 + 之前的 forward_varbest_tag_id = argmax(next_tag_var)bptrs_t.append(best_tag_id)viterbivars_t.append(next_tag_var[0][best_tag_id].view(1))# Now add in the emission scores, and assign forward_var to the set# of viterbi variables we just computedforward_var = (torch.cat(viterbivars_t) + feat).view(1, -1)  # 这个时候加上emission分数后 cat之后的格式转换， 然后再送入刚才 forward_varbackpointers.append(bptrs_t)# Transition to STOP_TAGterminal_var = forward_var + self.transitions[self.tag_to_ix[STOP_TAG]]best_tag_id = argmax(terminal_var)path_score = terminal_var[0][best_tag_id]# Follow the back pointers to decode the best path.best_path = [best_tag_id]for bptrs_t in reversed(backpointers):best_tag_id = bptrs_t[best_tag_id]best_path.append(best_tag_id)# Pop off the start tag (we dont want to return that to the caller)start = best_path.pop()assert start == self.tag_to_ix[START_TAG]  # Sanity checkbest_path.reverse()return path_score, best_pathdef neg_log_likelihood(self, sentence, tags):feats = self._get_lstm_features(sentence)forward_score = self._forward_alg(feats)gold_score = self._score_sentence(feats, tags)return forward_score - gold_scoredef forward(self, sentence):  # dont confuse this with _forward_alg above.# Get the emission scores from the BiLSTMlstm_feats = self._get_lstm_features(sentence)# Find the best path, given the features.score, tag_seq = self._viterbi_decode(lstm_feats)return score, tag_seq#####################################################################
# Run trainingSTART_TAG = "<START>"
STOP_TAG = "<STOP>"
EMBEDDING_DIM = 5
HIDDEN_DIM = 4# Make up some training data
training_data = [("the wall street journal reported today that apple corporation made money".split(),"B I I I O O O B I O O".split()
), ("georgia tech is a university in georgia".split(),"B I O O O O B".split()
)]word_to_ix = {}
for sentence, tags in training_data:for word in sentence:if word not in word_to_ix:word_to_ix[word] = len(word_to_ix)tag_to_ix = {"B": 0, "I": 1, "O": 2, START_TAG: 3, STOP_TAG: 4}model = BiLSTM_CRF(len(word_to_ix), tag_to_ix, EMBEDDING_DIM, HIDDEN_DIM)
optimizer = optim.SGD(model.parameters(), lr=0.01, weight_decay=1e-4)# Check predictions before training
with torch.no_grad():precheck_sent = prepare_sequence(training_data[0][0], word_to_ix)precheck_tags = torch.tensor([tag_to_ix[t] for t in training_data[0][1]], dtype=torch.long)print(precheck_tags)print(model(precheck_sent))# Make sure prepare_sequence from earlier in the LSTM section is loaded
for epoch in range(300):  # again, normally you would NOT do 300 epochs, it is toy datafor sentence, tags in training_data:# Step 1. Remember that Pytorch accumulates gradients.# We need to clear them out before each instancelog_sum_expmodel.zero_grad()# Step 2. Get our inputs ready for the network, that is,# turn them into Tensors of word indices.sentence_in = prepare_sequence(sentence, word_to_ix)targets = torch.tensor([tag_to_ix[t] for t in tags], dtype=torch.long)# Step 3. Run our forward pass.loss = model.neg_log_likelihood(sentence_in, targets)print(loss)# Step 4. Compute the loss, gradients, and update the parameters by# calling optimizer.step()loss.backward()optimizer.step()# Check predictions after training
with torch.no_grad():precheck_sent = prepare_sequence(training_data[0][0], word_to_ix)print(model(precheck_sent))
# We got it!

 

这篇关于pytorch bilstm crf NER的文章就介绍到这儿，希望我们推荐的文章对编程师们有所帮助！


http://www.chinasem.cn/article/506361。

相关文章

PyTorch使用教程之Tensor包详解

PyTorch使用教程之Tensor包详解

《PyTorch使用教程之Tensor包详解》这篇文章介绍了PyTorch中的张量（Tensor）数据结构,包括张量的数据类型、初始化、常用操作、属性等,张量是PyTorch框架中的核心数据结构,支持... 目录1、张量Tensor2、数据类型3、初始化（构造张量）4、常用操作5、常用属性5.1 存储（st
阅读更多...
Nn criterions don’t compute the gradient w.r.t. targets error「pytorch」 (debug笔记)

Nn criterions don’t compute the gradient w.r.t. targets error「pytorch」 (debug笔记)

Nn criterions don’t compute the gradient w.r.t. targets error「pytorch」 ##一、 缘由及解决方法 把这个pytorch-ddpg|github搬到jupyter notebook上运行时，出现错误Nn criterions don’t compute the gradient w.r.t. targets error。注：我用
阅读更多...
【超级干货】2天速成PyTorch深度学习入门教程，缓解研究生焦虑

【超级干货】2天速成PyTorch深度学习入门教程，缓解研究生焦虑

3、cnn基础 卷积神经网络 输入层 —输入图片矩阵 输入层一般是 RGB 图像或单通道的灰度图像，图片像素值在[0,255]，可以用矩阵表示图片 卷积层 —特征提取 人通过特征进行图像识别，根据左图直的笔画判断X，右图曲的笔画判断圆 卷积操作 激活层 —加强特征 池化层 —压缩数据 全连接层 —进行分类 输出层 —输出分类概率 4、基于LeNet
阅读更多...
pytorch torch.nn.functional.one_hot函数介绍

pytorch torch.nn.functional.one_hot函数介绍

torch.nn.functional.one_hot 是 PyTorch 中用于生成独热编码（one-hot encoding）张量的函数。独热编码是一种常用的编码方式，特别适用于分类任务或对离散的类别标签进行处理。该函数将整数张量的每个元素转换为一个独热向量。 函数签名 torch.nn.functional.one_hot(tensor, num_classes=-1) 参数 t
阅读更多...
pytorch计算网络参数量和Flops

pytorch计算网络参数量和Flops

from torchsummary import summarysummary(net, input_size=(3, 256, 256), batch_size=-1) 输出的参数是除以一百万（/1000000）M， from fvcore.nn import FlopCountAnalysisinputs = torch.randn(1, 3, 256, 256).cuda()fl
阅读更多...
Python(TensorFlow和PyTorch)两种显微镜成像重建算法模型(显微镜学)

Python(TensorFlow和PyTorch)两种显微镜成像重建算法模型(显微镜学)

🎯要点 🎯受激发射损耗显微镜算法模型：🖊恢复嘈杂二维和三维图像 | 🖊模型架构：恢复上下文信息和超分辨率图像 | 🖊使用嘈杂和高信噪比的图像训练模型 | 🖊准备半合成训练集 | 🖊优化沙邦尼尔损失和边缘损失 | 🖊使用峰值信噪比、归一化均方误差和多尺度结构相似性指数量化结果 | 🎯训练荧光显微镜模型和对抗网络图形转换模型 🍪语言内容分比 🍇Python图像归一化
阅读更多...
Pytorch环境搭建时的各种问题

Pytorch环境搭建时的各种问题

1 问题 1.一直soving environment，跳不出去。网络解决方案有：配置清华源，更新conda等，没起作用。2.下载完后，有3个要done的东西，最后那个exe开头的（可能吧），总是报错。网络解决方案有：用管理员权限打开prompt等，没起作用。3.有时候配置完源，安装包的时候显示什么https之类的东西，去c盘的用户那个文件夹里找到".condarc"文件把里面的网址都改成htt
阅读更多...
【PyTorch】使用容器(Containers)进行网络层管理(Module)

【PyTorch】使用容器(Containers)进行网络层管理(Module)

文章目录 前言一、Sequential二、ModuleList三、ModuleDict四、ParameterList & ParameterDict总结 前言 当深度学习模型逐渐变得复杂，在编写代码时便会遇到诸多麻烦，此时便需要Containers的帮助。Containers的作用是将一部分网络层模块化，从而更方便地管理和调用。本文介绍PyTorch库常用的nn.Sequen
阅读更多...
【python pytorch】Pytorch实现逻辑回归

【python pytorch】Pytorch实现逻辑回归

pytorch 逻辑回归学习demo: import torchimport torch.nn as nnimport torchvision.datasets as dsetsimport torchvision.transforms as transformsfrom torch.autograd import Variable# Hyper Parameters input_si
阅读更多...
【python pytorch】Pytorch 基础知识

【python pytorch】Pytorch 基础知识

包含知识点： 张量数学操作数理统计比较操作 #-*-coding:utf-8-*-import numpy as npnp.set_printoptions(suppress=True)import torch# 构造一个4*5 的矩阵z=torch.Tensor(4,5)print(z)# 两个矩阵进行加法操作y=torch.rand(4,5)print(z+y)# 另一种表示
阅读更多...

Copyright China编程 23002807@qq.com

沪ICP备2023033072号-2