引導式測驗 · 中等 等級

生產測驗中的數據品質監控

在推理之前檢測模式、新鮮度、缺失和列車服務問題,同時區分輸入品質和模型性能。

問題 1 的 8

What does a schema check detect in a live inference feed?

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. What does a schema check detect in a live inference feed?
    • Unexpected field names, types or structures
    • A guaranteed drop in model accuracy
    • A causal change in user behavior
    • The model's calibration error
  2. What does a freshness monitor detect?
    • A change in model weights
    • Data that have not been updated within the expected time
    • A label mismatch in a training batch
    • An unexpected prediction class
  3. How should an unseen categorical value be handled?
    • Drop the model silently
    • Map it randomly to a known category
    • Apply an explicit unknown-category policy and monitor its rate
    • Assume the value is always invalid
  4. Which mismatch is an example of train-serving skew?
    • A difference between two model architectures
    • A change in GPU memory usage
    • A delayed ground-truth label
    • A difference between how features are computed in training and inference
  5. Why can an overly strict distribution threshold cause production issues?
    • It guarantees detection of every defect
    • Legitimate seasonal changes may trigger false alarms or blocks
    • It prevents all schema changes
    • It improves model calibration automatically
  6. What does passing data-quality checks establish?
    • The feature relationships are unchanged
    • The model is fair across all groups
    • Inputs met the configured checks, not that predictions are accurate
    • The data are causally valid
  7. Where might latency-sensitive data checks run?
    • In a dashboard without storing results
    • All validation only after labels arrive
    • Only in the model-training notebook
    • Critical schema checks inline, with heavier analysis asynchronously or in batches
  8. What should a team do when a production data alert fires?
    • Trace the source and determine whether the change is expected before updating rules
    • Change the schema immediately to silence the alert
    • Retrain automatically without checking data
    • Delete the alert history