文本情感分类（三）：到底需不需要分词

深度学习是一种“端到端”的模型，所谓端到端就是能够将原始数据和标签输入，然后让模型自己完成一切过程-包括特征的提取、模型的学习。。而回顾我们做中文情感分类的过程，一般都是“分词——词向量——句向量(LSTM)——分类”这么几个步骤。虽然很多时候这种模型已经达到了state of art的效果，但是有些疑问还是需要进一步测试解决的。对于中文来说，字才是最低粒度的文字单位，因此从“端到端”的角度来看，应该将直接将句子以字的方式进行输入，而不是先将句子分好词。那到底有没有分词的必要性呢？本文测试比较了字one hot、字向量、词向量三者之间的效果。

模型测试
本文测试了三个模型，或者说是三套框架
1：one-hot 以字为单位，不分词，将每个句子截断为200字（不够则补齐空字符串），然后将句子以字one-hot的矩阵形式输入lstm模型中进行学习分类
2：one embedding:以字为单位，不分词，将每个句子截断为200字（不够则补齐空字符串），然后将句子以字-字向量的矩阵形式输入lstm模型中进行学习分类
3：word-embedding:以词的为单位，分词，将每隔句子截断为100词，然后将句子以词-词向量的矩阵形式输入到LSTM模型中进行学习分类。

其中所用的LSTM模型结构是类似的。意外的是，三个模型取得了类似的结果。

针对one-hot的理解
到底该不该舍弃one-hot？？？？？

模型1：one hot# -*- coding:utf-8 -*-'''one hot测试在GTX960上，约100s一轮经过90轮迭代，训练集准确率为96.60%，测试集准确率为89.21%Dropout不能用太多，否则信息损失太严重'''import numpy as npimport pandas as pdpos = pd.read_excel('pos.xls', header=None)pos['label'] = 1neg = pd.read_excel('neg.xls', header=None)neg['label'] = 0all_ = pos.append(neg, ignore_index=True)maxlen = 200 #截断字数min_count = 20 #出现次数少于该值的字扔掉。这是最简单的降维方法content = ''.join(all_[0])abc = pd.Series(list(content)).value_counts()abc = abc[abc >= min_count]abc[:] = range(len(abc))word_set = set(abc.index)def doc2num(s, maxlen):     s = [i for i in s if i in word_set]    s = s[:maxlen]    return list(abc[s])all_['doc2num'] = all_[0].apply(lambda s: doc2num(s, maxlen))#手动打乱数据#当然也可以把这部分加入到生成器中idx = range(len(all_))np.random.shuffle(idx)all_ = all_.loc[idx]#按keras的输入要求来生成数据x = np.array(list(all_['doc2num']))y = np.array(list(all_['label']))y = y.reshape((-1,1)) #调整标签形状from keras.utils import np_utilsfrom keras.models import Sequentialfrom keras.layers import Dense, Activation, Dropoutfrom keras.layers import LSTMimport syssys.setrecursionlimit(10000) #增大堆栈最大深度(递归深度)，据说默认为1000，报错#建立模型model = Sequential()model.add(LSTM(128, input_shape=(maxlen,len(abc)))) model.add(Dropout(0.5))model.add(Dense(1))model.add(Activation('sigmoid'))model.compile(loss='binary_crossentropy',              optimizer='rmsprop',              metrics=['accuracy'])#单个one hot矩阵的大小是maxlen*len(abc)的，非常消耗内存#为了方便低内存的PC进行测试，这里使用了生成器的方式来生成one hot矩阵#仅在调用时才生成one hot矩阵#可以通过减少batch_size来降低内存使用，但会相应地增加一定的训练时间batch_size = 128train_num = 15000#不足则补全0行gen_matrix = lambda z: np.vstack((np_utils.to_categorical(z, len(abc)), np.zeros((maxlen-len(z), len(abc)))))def data_generator(data, labels, batch_size):     batches = [range(batch_size*i, min(len(data), batch_size*(i+1))) for i in range(len(data)/batch_size+1)]    while True:        for i in batches:            xx = np.zeros((maxlen, len(abc)))            xx, yy = np.array(map(gen_matrix, data[i])), labels[i]            yield (xx, yy)model.fit_generator(data_generator(x[:train_num], y[:train_num], batch_size), samples_per_epoch=train_num, nb_epoch=30)model.evaluate_generator(data_generator(x[train_num:], y[train_num:], batch_size), val_samples=len(x[train_num:]))def predict_one(s): #单个句子的预测函数    s = gen_matrix(doc2num(s, maxlen))    s = s.reshape((1, s.shape[0], s.shape[1]))    return model.predict_classes(s, verbose=0)[0][0]模型2：one embedding# -*- coding:utf-8 -*-'''one embedding测试在GTX960上，36s一轮经过30轮迭代，训练集准确率为95.95%，测试集准确率为89.55%Dropout不能用太多，否则信息损失太严重'''import numpy as npimport pandas as pdpos = pd.read_excel('pos.xls', header=None)pos['label'] = 1neg = pd.read_excel('neg.xls', header=None)neg['label'] = 0all_ = pos.append(neg, ignore_index=True)maxlen = 200 #截断字数min_count = 20 #出现次数少于该值的字扔掉。这是最简单的降维方法content = ''.join(all_[0])abc = pd.Series(list(content)).value_counts()abc = abc[abc >= min_count]abc[:] = range(1, len(abc)+1)abc[''] = 0 #添加空字符串用来补全word_set = set(abc.index)def doc2num(s, maxlen):     s = [i for i in s if i in word_set]    s = s[:maxlen] + ['']*max(0, maxlen-len(s))    return list(abc[s])all_['doc2num'] = all_[0].apply(lambda s: doc2num(s, maxlen))#手动打乱数据idx = range(len(all_))np.random.shuffle(idx)all_ = all_.loc[idx]#按keras的输入要求来生成数据x = np.array(list(all_['doc2num']))y = np.array(list(all_['label']))y = y.reshape((-1,1)) #调整标签形状from keras.models import Sequentialfrom keras.layers import Dense, Activation, Dropout, Embeddingfrom keras.layers import LSTM#建立模型model = Sequential()model.add(Embedding(len(abc), 256, input_length=maxlen))model.add(LSTM(128)) model.add(Dropout(0.5))model.add(Dense(1))model.add(Activation('sigmoid'))model.compile(loss='binary_crossentropy',              optimizer='adam',              metrics=['accuracy'])batch_size = 128train_num = 15000model.fit(x[:train_num], y[:train_num], batch_size = batch_size, nb_epoch=30)model.evaluate(x[train_num:], y[train_num:], batch_size = batch_size)def predict_one(s): #单个句子的预测函数    s = np.array(doc2num(s, maxlen))    s = s.reshape((1, s.shape[0]))    return model.predict_classes(s, verbose=0)[0][0]模型3：word embedding# -*- coding:utf-8 -*-'''word embedding测试在GTX960上，18s一轮经过30轮迭代，训练集准确率为98.41%，测试集准确率为89.03%Dropout不能用太多，否则信息损失太严重'''import numpy as npimport pandas as pdimport jiebapos = pd.read_excel('pos.xls', header=None)pos['label'] = 1neg = pd.read_excel('neg.xls', header=None)neg['label'] = 0all_ = pos.append(neg, ignore_index=True)all_['words'] = all_[0].apply(lambda s: list(jieba.cut(s))) #调用结巴分词maxlen = 100 #截断词数min_count = 5 #出现次数少于该值的词扔掉。这是最简单的降维方法content = []for i in all_['words']:    content.extend(i)abc = pd.Series(content).value_counts()abc = abc[abc >= min_count]abc[:] = range(1, len(abc)+1)abc[''] = 0 #添加空字符串用来补全word_set = set(abc.index)def doc2num(s, maxlen):     s = [i for i in s if i in word_set]    s = s[:maxlen] + ['']*max(0, maxlen-len(s))    return list(abc[s])all_['doc2num'] = all_['words'].apply(lambda s: doc2num(s, maxlen))#手动打乱数据idx = range(len(all_))np.random.shuffle(idx)all_ = all_.loc[idx]#按keras的输入要求来生成数据x = np.array(list(all_['doc2num']))y = np.array(list(all_['label']))y = y.reshape((-1,1)) #调整标签形状from keras.models import Sequentialfrom keras.layers import Dense, Activation, Dropout, Embeddingfrom keras.layers import LSTM#建立模型model = Sequential()model.add(Embedding(len(abc), 256, input_length=maxlen))model.add(LSTM(128)) model.add(Dropout(0.5))model.add(Dense(1))model.add(Activation('sigmoid'))model.compile(loss='binary_crossentropy',              optimizer='adam',              metrics=['accuracy'])batch_size = 128train_num = 15000model.fit(x[:train_num], y[:train_num], batch_size = batch_size, nb_epoch=30)model.evaluate(x[train_num:], y[train_num:], batch_size = batch_size)def predict_one(s): #单个句子的预测函数    s = np.array(doc2num(list(jieba.cut(s)), maxlen))    s = s.reshape((1, s.shape[0]))    return model.predict_classes(s, verbose=0)[0][0]