提升机器算法

#!/usr/bin/env python3
# -*- coding:utf-8 -*-
"""
提升机器算法
LightGBM是个快速的，分布式的，高性能的基于决策树算法的梯度提升框架。
可用于排序，分类，回归以及很多其他的机器学习任务中。
"""
import multiprocessing
import numpy as np
import pandas as pd
from joblib import Parallel, delayed
import lightgbm as lgb
from sklearn.model_selection import cross_val_score, GridSearchCV, train_test_split
from wssc.app.config import DIM_DICT, DATASET


def create_bag_of_centroids(wordlist, dim_dict=DIM_DICT, num_centroids=len(DIM_DICT)):
    """
    向量袋
    :param wordlist:
    :param dim_dict:
    :param num_centroids:
    :return:
    """
    bag_of_centroids = np.zeros(num_centroids, dtype="float32")
    for word in wordlist:
        if dim_dict.__contains__(word):
            index = dim_dict.__getitem__(word)
            bag_of_centroids[index] += 1
    return bag_of_centroids


def main():
    NUM_CORES = multiprocessing.cpu_count()
    raw_data = pd.read_csv(DATASET % "result_1")
    raw_data["DIM"] = Parallel(n_jobs=NUM_CORES, verbose=10)(
        delayed(create_bag_of_centroids)(raw_data.iloc[index]["DIM"].split(";")) for index in range(len(raw_data)))
    X = np.array(raw_data["DIM"].values.tolist())
    Y = np.array(raw_data["C_ID"])
    seed = 7
    test_size = 0.22
    X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=test_size, random_state=seed)
    # lightgbm
    lgb_train = lgb.Dataset(X_train, y_train)  # 将数据保存到LightGBM二进制文件将使加载更快
    lgb_eval = lgb.Dataset(X_test, y_test, reference=lgb_train)  # 创建验证数据
    params = {
        'boosting_type': 'gbdt',
        'objective': 'multiclass',
        'num_class': 50,
        'metric': 'multi_error',
        'num_leaves': 300,
        'min_data_in_leaf': 100,
        'learning_rate': 0.01,
        'feature_fraction': 0.8,
        'bagging_fraction': 0.8,
        'bagging_freq': 5,
        'lambda_l1': 0.4,
        'lambda_l2': 0.5,
        'min_gain_to_split': 0.2,
        'verbose': 5,
        'is_unbalance': True
    }
    print('Start training...')
    # train
    gbm = lgb.train(params,
                    lgb_train,
                    num_boost_round=10000,
                    valid_sets=lgb_eval,
                    early_stopping_rounds=500)
    print('Start predicting...')
    # 预测数据集
    y_pred = gbm.predict(X_test, num_iteration=gbm.best_iteration)  # 如果在训练期间启用了早期停止，可以通过best_iteration方式从最佳迭代中获得预测
    pred = np.argmax(y_pred, axis=1)
    # 评估模型
    print((pred == y_test).mean())


if __name__ == '__main__':
    main()