에이전틱 앱에 대한 강력한 평가 기능 구현하기

5:16 - Generate synthetic data with makeSamples

// Synthetic data
  let prompt = Prompt("""
      Generate diverse range of book reviews and corresponding tags.
      Cover a wide range of genres, time periods, cultures, and
      reader personas. Do not repeat books already in the dataset.
      """)
  
  let dataset = Book.sampleBooks.map { book in
      ModelSample(prompt: book.review, expected: BookTags(tags: book.tags))
  }
  
  let targetCount = 100
  var expandedDataset = dataset

  for try await sample in dataset.makeSamples(prompt, targetCount: targetCount) {
      expandedDataset.append(sample)
      print("Generated \(expandedDataset.count) samples so far.")
  }

  2. Configure a custom SampleGenerator — slides 30–43
  
  // Define your own configuration
  let generator = SampleGenerator<ModelSample<BookTags>>(
      prompt,
      samples: dataset,
      targetCount: targetCount,
      sessionProvider: {
          LanguageModelSession( 
              model: PrivateCloudComputeLanguageModel(),
              instructions: """
                  You are a synthetic data generator for a book-tracking app's evaluation suite.
                  Your job is to produce realistic, diverse book entries that will stress-test
                  a tagging system.

                  Rules:
                  - Review must be at least 100 characters long.
                  - Review should cover a mix of genre, mood/tone, and themes.
                  - Reviews should vary in length.
                  - Create between 3 and 8 tags.
                  - Tags must be lowercase.
                  """ 
          )
      }
  )

5:53 - Configure a custom SampleGenerator

// Define your own configuration
  let generator = SampleGenerator<ModelSample<BookTags>>(
      prompt,
      samples: dataset,
      targetCount: targetCount,
      sessionProvider: {
          LanguageModelSession( 
              model: PrivateCloudComputeLanguageModel(),
              instructions: """
                  You are a synthetic data generator for a book-tracking app's evaluation suite.
                  Your job is to produce realistic, diverse book entries that will stress-test
                  a tagging system.

                  Rules:
                  - Review must be at least 100 characters long.
                  - Review should cover a mix of genre, mood/tone, and themes.
                  - Reviews should vary in length.
                  - Create between 3 and 8 tags.
                  - Tags must be lowercase.
                  """ 
          )
      }
  )

10:37 - Validate generated samples

// Define validation metrics
  validator: { sample in
      guard let book = sample.expected else { return false }

      // Review must be at least 100 characters
      guard sample.promptDescription.count >= 100 else { return false }

      // Must have between 3 and 8 tags
      guard (3...8).contains(book.tags.count) else { return false }

      // All tags must be lowercase
      guard book.tags.allSatisfy({ $0 == $0.lowercased() }) else { return false }

      return true
  }

10:58 - Access valid and invalid results

// Accessing results
  for try await sample in generator.run() {
      // During iteration
      expandedDataset.append(sample)
  }

  // After iteration
  let allSamples = await generator.samples
  let invalidSamples = await generator.invalidSamples
  
  print("Generated \(allSamples.count) new samples. Total: \(expandedDataset.count)")

15:30 - Define a tool's Generable argument

@Generable
  struct SearchBooksArguments {
      @Guide(description: "A freeform search term to match against titles, reviews, or tags")
      var query: String?
  
      @Guide(description: "Filter results to books with this specific tag")
      var tag: String?

      @Guide(description: "Filter results by mood")
      var mood: String?

      @Guide(description: "Filter results by genre")
      var genre: String?

      @Guide(description: "Maximum number of results to return. Defaults to 5.")
      var limit: Int? 
  }

16:37 - A basic trajectory expectation

// "Find books tagged gothic"
  TrajectoryExpectation(
      unordered: [
          ToolExpectation(
              "searchBooks",
              arguments: [
                  .exact(argumentName: "tag", value: .string("gothic"))
              ]
          )
      ]
  )

17:07 - Match arguments by intent (naturalLanguage)

// "Find something cheerful"
  TrajectoryExpectation(
      "searchBooks",
      arguments: [
          .naturalLanguage(
              argumentName: "mood",
              criteria: "Should relate to uplifting, hopeful, or positive feelings"
          )
      ]
  )
  Other matchers available: .contains, .oneOf, .pattern, .range, and more.

17:34 - Expect tool calls in order

// "Find gothic books and show details on the first"
  TrajectoryExpectation(
      ordered: [
          ToolExpectation(
              "searchBooks",
              arguments: [
                  .exact(argumentName: "tag", value: .string("gothic"))
              ]
          ),
          ToolExpectation(
              "getBookDetails",
              arguments: [
                  .keyOnly(argumentName: "bookId")
              ]
          )
      ]
  )

17:55 - Disallow specific tool calls

// "Show only sci-fi books. Don't look for similar ones."
  TrajectoryExpectation(
      unordered: [
          ToolExpectation(
              "searchBooks",
              arguments: [
                  .naturalLanguage(
                      argumentName: "genre",
                      criteria: "Should refer to science fiction")
              ]
          )
      ],
      disallowed: [
          ToolExpectation("findSimilarBooks")
      ]
  )

18:14 - Build a tool call evaluation

// Tool call evaluations
  let samples = SampleArrayLoader(samples: [
      ModelSample(
          prompt: "Find all the books tagged with 'gothic'.",
          instructions: "Help the user explore their book collection.",
          expectations: TrajectoryExpectation(  )
      )
  ])

  struct BookLibraryToolCallEval: Evaluation {
      var dataset = samples

      let pass = Metric("All Passed")
      let percent = Metric("Percentage Passed")

      var evaluators: Evaluators { 
          ToolCallEvaluator(allPass: pass, percentagePass: percent)
      }
  }

19:20 - Synthesize tool-evaluation samples

// Tool call evaluations
  let prompt = Prompt("""
      Generate diverse user queries for a personal book library assistant.
      Each sample needs a prompt (what the user says), and a trajectory
      expectation describing which tools should be called and in what order.
      """)

  let instructions = """
      AVAILABLE TOOLS:
      - searchBooks(query?, tag?, mood?, genre?, limit?): search the library
      - getBookDetails(bookId): full details for one book
      - findSimilarBooks(bookId, maxResults?): find books sharing tags
      ORDER REQUIREMENTS:
      - searchBooks must comes before getBookDetails or findSimilarBooks
      - Use TrajectoryExpectation(ordered:) when sequence matters, else (unordered:)
      USE THESE ARGUMENT MATCHERS:
      - .exact for precise values, .naturalLanguage for fuzzy matching
      - .keyOnly when any value is acceptable, .range for numeric constraints
      - .contains/.hasPrefix/.hasSuffix for partial string matching
      """

19:51 - Validate tool-evaluation samples

// Tool call evaluations
  validator: { sample in
      // Must have expectations defined
      guard sample.output.expectations != nil else { return false }

      let expectations = sample.output.expectations!

      // Must reference at least one tool
      let totalExpectations = expectations.ordered.count + expectations.unordered.count
      guard totalExpectations > 0 else { return false }

      // All tool names must be from the valid set
      let validTools: Set<String> = ["searchBooks", "getBookDetails", "findSimilarBooks"]
      let allExpectations = expectations.ordered + expectations.unordered + expectations.disallowed
      for expectation in allExpectations {
          guard validTools.contains(expectation.name) else { return false }
      }
  
      return true
  }

  ---