dask - 使用 dask_xgboost 训练方法时,Dask 挂起
问题描述
我正在尝试从http://ml.dask.org/examples/xgboost.html的 dask-ml 文档中重现 dask xgboost 示例。不幸的是,Dask 似乎没有完成培训,我很难追查错误和警告的含义。这是我的代码:
def main():
cluster = LocalCluster()
dask_client = Client(cluster)
x, y = make_classification(n_samples=1000, n_features=10, chunks=10, n_informative=4, random_state=0)
X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=0.15)
print("classification training...")
params = {'objective': 'binary:logistic',
'max_depth': 4,
'eta': 0.01,
'subsample': 0.5,
'min_child_weight': 0.5}
bst = dask_xgboost.train(dask_client, params, X_train, y_train, num_boost_round=10)
dask_client.close()
if __name__ == '__main__':
main()
这是输出:
classification training...
Socket Poll Error:WSAError-code=10022
Socket Poll Error:WSAError-code=10022
Socket Poll Error:WSAError-code=10022
Socket Poll Error:WSAError-code=10022
distributed.nanny - WARNING - Worker process 12376 exited with status 4294967295
distributed.nanny - WARNING - Worker process 3540 exited with status 4294967295
distributed.nanny - WARNING - Worker process 9424 exited with status 4294967295
distributed.nanny - WARNING - Worker process 12952 exited with status 4294967295
distributed.nanny - WARNING - Restarting worker
distributed.nanny - WARNING - Restarting worker
distributed.nanny - WARNING - Restarting worker
distributed.nanny - WARNING - Restarting worker
有什么想法吗?
解决方案
推荐阅读
- mysql - MySQL 查询 (SELECT) 的问题
- firebase - 在颤振应用程序中更新日期
- r - 如何根据条件更改列表中多个数据框中的列名称?
- reactjs - 使用其他钩子时 usePrevious 可靠吗?
- c++ - 我无法让我的 MergeSort 实现运行
- javascript - 使用 AJAX 在传单上重新加载热图
- java - 如何在 Android 上为 Highcharts 正确配置 DateTime 轴?
- hibernate - 比较 dateCreated 的查询不适用于在 Grails 3.3.10 的测试事务中创建的测试数据
- matplotlib - 从 matplotlib.animation 加载“FuncAnimation”时遇到问题
- flutter - 我该如何解决颤振错误