[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-03-06。"],[[["This page explains how AI Platform Training preprocesses tabular data for built-in algorithms, detailing the requirements and limitations for input data."],["AI Platform Training analyzes tabular data column by column to determine data type, treatment, and statistics, saving this information in a `metadata.json` file."],["Data transformation involves splitting datasets, removing rows with excessive missing values, and filling in the rest of the missing values, using the mean for numerical columns and zeros for categorical ones in XGBoost."],["Input data must be a UTF-8 encoded CSV file, without a header row, and with the target column as the first column."],["The preprocessing operations vary depending on the ML framework, with TensorFlow based models performing these transformations inside the TensorFlow graph, while XGBoost built-in algorithms perform these transformations outside of the TensorFlow graph."]]],[]]