# import re
# import torch
# import torch.nn as nn
# import jieba
# import pandas as pd
# # from torchtext import data
#
# class LSTMNet(nn.Module):
#   def __init__(self,vocab_size,embedding_dim,hidden_dim,layer_dim,output_dim):
#     super(LSTMNet,self).__init__()
#     self.hidden_dim= hidden_dim
#     self.layer_dim = layer_dim
#     self.embedding = nn.Embedding(vocab_size,embedding_dim)
#     # LSTM+全连接
#     self.lstm = nn.LSTM(embedding_dim,hidden_dim,layer_dim,
#               batch_first=True)
#     self.fcl= nn.Linear(hidden_dim,output_dim)
#   def forward(self,x):
#     embeds = self.embedding(x)
#     r_out,(h_n,h_c)=self.lstm(embeds,None)
#     out = self.fcl(r_out[:,-1,:])
#     return out
#
# def Chinese_pre(text_data,stopwords):
#   # 字母转化为小写， 去掉数字
#   text_data = text_data.lower()
#   text_data = re.sub("\d+","",text_data)
#   # 分词，使用精确模式
#   text_data = list(jieba.cut(text_data,cut_all = False))
#   # 去除停用词和多余空格
#   text_data = [word.strip() for word in text_data if word not in stopwords]
#   # 处理后的词语使用空格连接为字符串
#   text_data = " ".join(text_data)
#   return text_data
#
# def TexttoLable(textdata):
#   # 将输入文本转为tensor
#   # 首先对文本进行分词
#   from nltk.corpus import stopwords
#   import nltk
#   nltk.download('stopwords')
#   words = stopwords.words('english')
#   stopwords = set()
#   with open("stop.txt",encoding="utf-8") as infile:
#       for line in infile:
#           line = line.rstrip('\n')
#           if line:
#               stopwords.add(line.lower())
#   for i in words:
#     stopwords.add(i)
#   textdata=Chinese_pre(textdata,stopwords)
#
#   data1=[]
#   for i in range(128):
#     data1.append(textdata)
#   df = pd.DataFrame({'cutword':data1})
#
#   df.to_csv("tmp.csv")
#
#   mytokenize = lambda x:x.split()
#   from torchtext.legacy import data
#   TEXT = data.Field(sequential = True,tokenize = mytokenize,
#             include_lengths=True,use_vocab=True,
#             batch_first=True,fix_length=40)
#
#   LABEL = data.Field(sequential =False,use_vocab=False,
#             pad_token=None,unk_token=None)
#   # 对所有读取的数据集的列进行处理
#   text_data_fields = [
#     ("labelcode",LABEL),
#     ("cutword",TEXT)
#   ]
#   # 读取数据
#   # 读取数据
#   traindata,valdata,testdata = data.TabularDataset.splits(
#       path="./",format="csv",train="tmp.csv",fields = text_data_fields,
#       validation = "tmp.csv",
#       test ="tmp.csv",skip_header=True
#   )
#
#   em = testdata.examples[0]
#   TEXT.build_vocab(traindata,max_size=100,vectors=None)
#
#   # 定义一个迭代器，将类似长度的示例一起批处理
#   BATCH_SIZE=128
#   test_iter = data.BucketIterator(testdata,batch_size=128)
#
#
#   vocab_size=len(TEXT.vocab)
#   embedding_dim=50
#   hidden_dim=256
#   layer_dim=1
#   output_dim=4
#   lstmmodel = LSTMNet(vocab_size, embedding_dim, hidden_dim, layer_dim, output_dim)
#
#   res=0
#   model = torch.load("model.pkl")
#   for step,batch in enumerate(test_iter):
#         textfinal = batch.cutword[0]
#         out = model(textfinal)
#         pre_lab = torch.argmax(out,1)
#         res = pre_lab[0]
#   print(res.numpy())
#
# TexttoLable("萝卜云服交流群等3个会话	")