情景

在使用 pytorch 1.0 和 RTX 3060 Laptop 训练 GIN 的时候出现了这个错误

原因

pytorch 1.0 对应的 cuda 10.0 版本太低了，不支持新的 GPU 架构

解决

最终使用了更新版本的 pytorch

本栏目推荐文章
• 2021-2022 ICPC Northwestern European Regional Programming Contest (NWERC 2021)
• 安装npm install报错npm ERR! code ETIMEDOUT npm ERR! errno ETIMEDOUT npm ERR! network request to https://registry.npmjs.org/webpack-subresource-integrity failed, reason
• 2023 United Kingdom and Ireland Programming Contest (UKIEPC 2023)
• 2022 Jiangsu Collegiate Programming Contest
• scp 出现 subsystem request failed on channel 0 scp: Connection closed 的解决方法
• Docker 数据库连接见解异常 SQLSTATE[HY000] [2002] php_network_getaddresses: getaddrinfo failed: Try again
• cdn引入vue后报错无法识路径 Uncaught TypeError: Failed to resolve module specifier "vue". Relative references must start with either "/", "./", or "../".
• 2021 Jiangsu Collegiate Programming Contest
• [Vue warn]: Component provided template option but runtime compilation is not supported in this build of Vue. Configure your bundler to alias "vue" to "vue/dist/vue.esm-bundler.js".
• SpringBoot配置报错：Description: Failed to configure a DataSource: 'url' attribute is not specified and no embedded datasource could be configured. Reason: Failed to determine a suitable driver class

runtime program execute cublas failedruntime program execute cublas codehaus execute failed goal mapredtask execution failed return hive mapredlocaltask execution failed xmlhttprequest failed 39 execute cublas runtime execute execution executable