引導式測驗 · 中等 等級

資料驗證與遠大的期望測驗

學習 GX Core 概念,並將驗證結果與資料修復、策略判斷和資料集品質證明區分開來。

相關引導路徑數據驗證充滿期待
問題 1 的 8

How does GX Core define an Expectation?

預覽每個問題8 問題

當您參加測驗時,就會出現答案和解釋。

  1. How does GX Core define an Expectation?
    • As a guarantee that every downstream model is fair
    • As a command that automatically fixes all rows
    • As a verifiable assertion about the acceptable state of data
    • As a database backup
  2. What does a Validation Definition connect in the current GX Core workflow?
    • A user account and a billing plan
    • A model and a fairness certificate
    • A dashboard and a database password
    • A Batch Definition and an Expectation Suite
  3. What can a Checkpoint do?
    • Automatically guarantee that input data are correct
    • Run one or more validations and then configured Actions
    • Repair every failed row without instructions
    • Replace data governance
  4. If all Expectations pass, what has been established?
    • The tested batch satisfied the configured rules, not every possible quality requirement
    • The dataset is unbiased and complete
    • The business rules are necessarily correct
    • All future batches will pass
  5. Do GX validations automatically clean or repair bad data by default?
    • Yes, all failures are silently rewritten
    • Yes, but only in Data Docs
    • No; they report validation results, and teams decide how to respond
    • No, because GX cannot detect failures
  6. What role can Checkpoint Actions serve?
    • They decide whether a business threshold is ethical
    • They can run configured steps such as updating Data Docs or sending notifications
    • They create guaranteed model accuracy
    • They always halt production automatically
  7. Why should teams use docs for their installed GX Core version?
    • All versions use identical interfaces
    • Documentation determines the data values
    • Older tutorials automatically update code
    • GX APIs and workflows have changed, including around the 1.0 release
  8. Which conclusion is warranted by a passing data validation?
    • The checked data met the explicit tests; separate review may still be needed
    • The dataset is safe for every purpose
    • No data drift is possible
    • All missing requirements have been discovered