赛题目标:
建立模型通过长文本数据正文,预测文本对应的类别
数据准备:
train_set.csv 用于模型训练
test_set.csv 用于模型预测
读取前5000条数据:
import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
my_data = pd.read_csv('../datasets/train_set.csv', nrows = 5000)
查看缺失情况、文本类别数量:
my_data.isnull().sum()
my_data['class'].value_counts()
发现数据中无缺失值,文本类别共分为19类
拆分训练集和验证集:
X, y = my_data.ix[:,:-1], my_data.ix[:,-1]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.3, random_state = 42)
train_data = pd.DataFrame(np.column_stack((X_train, y_train)), columns = my_data.columns)
test_data = pd.DataFrame(np.column_stack((X_test, y_test)), columns = my_data.columns)
拆分后训练集和验证集大小分别为3500和1500:
print(train_data.shape, test_data.shape)
根据赛题提供数据可得,虽然赛题为文本处理,但数据都是脱敏后的,不需要通过jieba等做切词处理,直接对数据建模即可。