architecture/network/stats

Raised when network test helpers receive a missing or empty evaluation set.

architecture/network/stats/network.stats.errors.ts

NetworkStatsTestSampleInputSizeMismatchError

Raised when a test sample input vector does not match network input width.

NetworkStatsTestSampleOutputSizeMismatchError

Raised when a test sample output vector does not match network output width.

NetworkStatsTestSetValidationError

Raised when network test helpers receive a missing or empty evaluation set.

architecture/network/stats/network.stats.utils.ts

Network statistics accessors.

Currently exposes a single helper for retrieving the most recent regularization / stochasticity metrics snapshot recorded during training or evaluation. The internal _lastStats field on the Network instance is read through the local NetworkStatsProps bridge and is expected to be populated elsewhere in the training loop with values such as:

• l1Penalty, l2Penalty
• dropoutApplied (fraction of units dropped last pass)
• weightNoiseStd (effective std dev used if noise injected)
• sparsityRatio, prunedConnections
• custom user extensions (the object stays intentionally open for experimentation)

Design decision: We return a deep copy to prevent external mutation of internal accounting state. If the object is large and copying becomes a bottleneck, future versions could offer a freeze option or incremental diff interface.

getRegularizationStats

getRegularizationStats(): Record<string, unknown> | null

Obtain the last recorded regularization / stochastic statistics snapshot.

Returns a defensive deep copy so callers can inspect metrics without risking mutation of the internal _lastStats object maintained by the training loop (e.g., during pruning, dropout, or noise scheduling updates).

Returns: A deep-cloned stats object or null if no stats have been recorded yet.

networkStatsUtils

Default export bundle for the network statistics utilities chapter.

Bundles getRegularizationStats so the network facade can bind it as a method without importing it individually.

testNetwork

testNetwork(
  set: TestSample[],
  cost: CostFunction | undefined,
): TestNetworkResult

Evaluate a dataset and return average error and elapsed time.

Parameters:

• this - Bound network instance.
• set - Evaluation samples.
• cost - Optional cost function override.

Returns: Mean error and evaluation duration.

architecture/network/stats/network.stats.test.utils.ts

createTestResult

createTestResult(
  cumulativeError: number,
  sampleCount: number,
  startTime: number,
): TestNetworkResult

Build the final test result payload.

Parameters:

• cumulativeError - Cumulative sample error.
• sampleCount - Number of evaluated samples.
• startTime - Evaluation start timestamp.

Returns: Mean error and elapsed duration.

disableDropoutForTesting

disableDropoutForTesting(
  network: default,
): number

Disable dropout while preserving previous runtime dropout value.

Parameters:

• network - Bound network instance.

Returns: Previous dropout value.

evaluateSamples

evaluateSamples(
  network: default,
  testSet: TestSample[],
  costFunction: CostFunction,
): number

Evaluate all test samples and accumulate total cost.

Parameters:

• network - Bound network instance.
• testSet - Evaluation sample set.
• costFunction - Cost function used for scoring.

Returns: Cumulative error across all samples.

evaluateSingleSample

evaluateSingleSample(
  network: default,
  sample: TestSample,
  costFunction: CostFunction,
): number

Evaluate a single sample and return its cost.

Parameters:

• network - Bound network instance.
• sample - Evaluation sample.
• costFunction - Cost function used for scoring.

Returns: Error for the sample.

resetHiddenMasks

resetHiddenMasks(
  network: default,
): void

Force hidden-node masks to active state for deterministic testing.

Parameters:

• network - Bound network instance.

resolveCostFunction

resolveCostFunction(
  cost: CostFunction | undefined,
): CostFunction

Resolve evaluation cost function with a stable default.

Parameters:

• cost - Optional cost override.

Returns: Cost function used for test evaluation.

restoreDropout

restoreDropout(
  network: default,
  previousDropout: number,
): void

Restore dropout value after test evaluation.

Parameters:

• network - Bound network instance.
• previousDropout - Dropout value to restore.

testNetwork

testNetwork(
  set: TestSample[],
  cost: CostFunction | undefined,
): TestNetworkResult

Evaluate a dataset and return average error and elapsed time.

Parameters:

• this - Bound network instance.
• set - Evaluation samples.
• cost - Optional cost function override.

Returns: Mean error and evaluation duration.

validateAllSampleDimensions

validateAllSampleDimensions(
  network: default,
  testSet: TestSample[],
): void

Validate input and output dimensions for every sample.

Parameters:

• network - Bound network instance.
• testSet - Evaluation sample set.

validateSampleInputDimensions

validateSampleInputDimensions(
  network: default,
  sample: TestSample,
): void

Validate one sample input vector size.

Parameters:

• network - Bound network instance.
• sample - Evaluation sample.

validateSampleOutputDimensions

validateSampleOutputDimensions(
  network: default,
  sample: TestSample,
): void

Validate one sample output vector size.

Parameters:

• network - Bound network instance.
• sample - Evaluation sample.

validateTestSet

validateTestSet(
  network: default,
  testSet: TestSample[],
): void

Validate that the evaluation set exists and each sample matches network dimensions.

Parameters:

• network - Bound network instance.
• testSet - Evaluation sample set.

validateTestSetPresence

validateTestSetPresence(
  testSet: TestSample[],
): void

Validate that the test set is a non-empty array.

Parameters:

• testSet - Evaluation sample set.