Hanamaru 日本語にほんご × ML/AI
← All topics

Core Machine Learning Topics

Model Evaluation

モデル評価ひょうか

Model evaluation measures whether a model meets its intended purpose. An evaluation metric quantifies performance, while cross-validation estimates performance across multiple data partitions.

Japanese terms

  1. Model evaluation — モデル評価ひょうか: The systematic measurement of how well a model performs for its intended purpose.
  2. Evaluation metric — 評価指標ひょうかしひょう: A numerical measure used to assess a model’s predictions or behavior.
  3. Cross-validation — 交差検証こうさけんしょう: A resampling method that evaluates a model across multiple training and validation partitions.

Training, validation, and testing establishes evaluation data, and AI safety and evaluation extends measurement to risks and robustness.