python cnn代码详解 keras_Textcnn多分类（keras实现）示例详细信息,textcnn,实例,详解...

这是一个非常经典的代码，我增加了一定的注释和保存模型的代码

使用的语料库和数据集都是网站公开数据(百度直接搜索名字都能搜到，下面提供下载地址)

from __future__ import print_function

import os

import sys

import numpy as np

from keras.preprocessing.text import Tokenizer

from keras.preprocessing.sequence import pad_sequences

from keras.utils import to_categorical

from keras.layers import Dense, Input, GlobalMaxPooling1D

from keras.layers import Conv1D, MaxPooling1D, Embedding

from keras.models import Model

from keras.initializers import Constant

from keras.callbacks import ModelCheckpoint

BASE_DIR = ''

#glove模型路径

GLOVE_DIR = os.path.join(BASE_DIR, 'glove.6B')

#文本语料路径

TEXT_DATA_DIR = os.path.join(BASE_DIR, '20_newsgroup')

MAX_SEQUENCE_LENGTH = 1000

MAX_NUM_WORDS = 20000

EMBEDDING_DIM = 100

VALIDATION_SPLIT = 0.2

# first, build index mapping words in the embeddings set

# to their embedding vector

#1.准备glove词向量和它们对应的字典映射

print('Indexing word vectors.')

#我们从GloVe文件中解析出每个词和它所对应的词向量，并用字典的方式存储

embeddings_index = {}

with open(os.path.join(GLOVE_DIR, 'glove.6B.100d.txt')) as f:

for line in f:

word, coefs