不是phd的phd

论文名：AliCoCo: Alibaba E-commerce Cognitive Concept Net引言基本电商的图谱建设都是Category-Property-Value形式，Category是一颗树，其中Property就是 Category的叶子 下的属性。问题在于，用户通常不记得一个item的具体名字，而是只记得item的类别或者item其他相关信息。分类学taxonomy人工定义好taxonomy（分类学），domain有20种，主要的是Category, Brand, Colo

import reprint(re.findall("防治.{1,2}动脉硬化","防治动脉硬化"))print(re.findall("防治.{1,2}动脉硬化","防治高血压动脉硬化"))print(re.findall("防治.{1,3}动脉硬化","防治高血压动脉硬化"))print(re.findall("防治.{1,4}动脉硬化","防治高血压动脉硬化"))prin...

DataFrame.columns.values.tolist()

hw.c的内容：#include <iostream>#include <stdint.h>#include <xgboost/c_api.h>using namespace std;int main(int argc, char const *argv[]){cout << "Hello World!" << endl;...

XGDMatrixCreateFromMat传入的数据有正特大或负特大的值

https://stackoverflow.com/questions/37167795

就是训练了比如1000 iteration，但第666次是最好的模型，怎么用这第666 iteration的而不是最后一次iteration的模型？ 在CPP API https://github.com/dmlc/xgboost/blob/master/include/xgboost/c_api.h 找了半天，就XGBoosterEvalOneIter接近解决这个问题， 尝试后，XGBoo...

xgboost 0.72版本 将xgboost的CMakeList.txt里的openmp相关的删了 加锁pthread_mutex_lock(&lock_); XGBoosterPredict(h_booster, h_test, 0, 0, &out_len, &f);pthread_mutex_unlock(&lock_);...

https://github.com/dmlc/treelite

two-text-match这个任务，如果我把"谁写的","谁的著作"这两个词看为同义词，则我强行指定 分别出现这两个词的 这对句子match程度为1如果我把"古代流放", "流放"不看成同义词，则我强行指定 分别出现这两个词的 这对句子match程度为-1然后add_data = train_x[train_x[&quo

pd.DataFrame({"column0":column0_data}).shape[0]

from sklearn import metrics#valid_predict是0到1的值，未进行设定阈值划分为0和1precision, recall, thresholds = metrics.precision_recall_curve(valid_label, valid_predict)all_f1 = []for i in range(len(thresholds)): ...

#省略……from sklearn.model_selection import train_test_splittrain_x, test_x, train_y, test_y = train_test_split(feature_matrix, labels, random_state=0)import xgboost as xgbdtrain=xgb.DMatrix(train...

准确率 - accuracy 精确率 - precision每条的统计由下图 然后

df_train = pd.read_csv('data/train.csv',encoding='utf-8')df_train.to_csv("train_preprocessed.csv",encoding='utf-8',index=False)设置index=False

xgb_trained_model = xgb.train(params, train_xgb_split, 500, watch_list, early_stopping_rounds=50, verbose_eval=10)importance = xgb_trained_model.get_fscore()temp1 = []temp2 = []for k in importanc...

[240] train-logloss:0.263565 valid-logloss:0.392514[250] train-logloss:0.261231 valid-logloss:0.392377[260] train-logloss:0.257999 valid-logloss:0.392149[270] train-logloss:0.254814 val...

add_data[add_data!=0]=0add_data["one_column"]=1

def my_test(a, b): return a + bdf['value'] = df.apply(lambda row: my_test(row['c1'], row['c2']), axis=1)

print(pd.DataFrame({ 'column': feature_names, 'importance': lgb_trained_model.feature_importance(), }).sort_values(by='importance'))