XGB模型训练报错 terminate called after throwing an instance of ‘std::bad

背景：需要做XGB模型增量训练，但是因为一些原因没有得到原来的XGB模型，只有dump文件。

XGB模型的dump文件是它的树结构存储为一个相对人类好理解的模型文件，但并不能被重新加载进XGB进行增量训练。

嗯，说是那么说，我寻思试一下，看看最新版的XGB有没有增加这个功能，只载入dump文件就可以进行XGB增量训练。

一跑起来没报错，我寻思可以呢。

一开始加载数据啥的都没问题，但是在训练的时候报错：terminate called after throwing an instance of 'std::bad_alloc' what():，还把环境整崩了。

搜了一下，都说这是内存炸了导致的报错，但是我看了一下监控，内存完全没占满。

到这我还一直以为可能是我训练数据太大，或者是后台起的训练进程太多才导致的，完全没想过是不是dump文件的问题。

是在解决不了了，去洗了个澡，突然间就想到了，可能是dump文件的问题。

实验了一下，果然是，XGB模型增量训练只能在save_model保存的模型文件上训练，不能在dump_model的dump文件上训练，它不报解析错误反而报了一个内存错误。

实验代码：

import pandas as pd
import xgboost as xgb
import matplotlib.pyplot as pltfrom xgboost import plot_tree
from xgboost import XGBClassifierdata = pd.read_csv("../input/default-of-credit-card-clients-dataset/UCI_Credit_Card.csv")
X = data.iloc[:,1:6]
y = data.iloc[:,24]data = xgb.DMatrix(X, y)
xgb_params = {}
model = xgb.train(xgb_params, data)
print(model.get_dump()) model.save_model("model.pickle")
model.dump_model("model.dump")xgb_params = {'process_type':'update', 'updater':'refresh', 'refresh_leaf':True}
model_new = xgb.train(xgb_params, data, xgb_model="./model.pickle")xgb_params = {'process_type':'update', 'updater':'refresh', 'refresh_leaf':True}
model_new = xgb.train(xgb_params, data, xgb_model="./model.dump")   # 报错

XGB模型训练报错 terminate called after throwing an instance of ‘std::bad_alloc‘ what()相关推荐

c++运行时报错terminate called after throwing an instance of ‘std::bad_alloc‘
程序运行时发生错误 terminate called after throwing an instance of 'std::bad_alloc'what(): std::bad_alloc 代码类似 ...
报错信息 terminate called after throwing an instance of ‘std::cad_alloc‘ what():std::bad_alloc
最近做了一道bfs算法的程序题输入测试数据后卡在运行框中然后会产生一个错误信息第一次见到这样的报错于是记录一下解决过程 terminate called after throwing an in ...
terminate called after throwing an instance of 'std::out_of_range' what(): basic_string::substr
运行时报错: terminate called after throwing an instance of 'std::out_of_range' what(): basic_string::sub ...
Linux运行python文件出现以下错误：terminate called after throwing an instance of ‘std::runtime_error‘
Linux运行python文件出现以下错误:terminate called after throwing an instance of 'std::runtime_error'
terminate called after throwing an instance of ‘std::runtime_error‘ what(): locale::facet::_S_cre
在运行程序时出错: terminate called after throwing an instance of 'std::runtime_error' what(): locale::fac ...
错误：【terminate called after throwing an instance of 'std::logic_err】
自己写代码的时候,出现了一下提示 terminate called after throwing an instance of 'std::logic_error'what(): basic_stri ...
terminate called after throwing an instance of ‘std::runtime_error‘
terminate called after throwing an instance of 'std::runtime_error' what(): random_device::random_d ...
Centos7 编译C++项目错误解决： terminate called after throwing an instance of ‘std::regex_error‘
文章目录 1 问题原因 1.1 确保编译器支持std::regex 2 解决方法 3 Centos7升级gcc 3.1 安装centos-release-scl 3.2 安装devtoolset 3. ...
keras模型训练报错： AttributeError: 'list' object has no attribute 'set_model'
错误信息: AttributeError: 'list' object has no attribute 'set_model' 根据错误提示:错误代码位于如下 model.fit(X_train, ...

XGB模型训练报错 terminate called after throwing an instance of ‘std::bad_alloc‘ what()

XGB模型训练报错 terminate called after throwing an instance of ‘std::bad_alloc‘ what()相关推荐

最新文章

热门文章