afe.apis.statistic

Source: afe/apis/statistic.py

Back to index

Analysis of statistics on tensors.

Imports

• afe.apis.compilation_job_base.Tensor
• afe.core.evaluate_networks.checked_zip
• afe.driver.statistic.Statistic
• dataclasses
• numpy as np
• typing.Any
• typing.Callable
• typing.Generic
• typing.List
• typing.Tuple
• typing.Type
• typing.TypeVar

Constants

• Metric (line 28) [default/value Callable[[Tensor, Tensor], float]]

Functions

• equality(x: Tensor, y: Tensor) -> float (line 31): Equality as a distance metric.

  Parameters:

  • x: type Tensor
  • y: type Tensor

  Returns: float

• mean_float(x: Tensor, y: Tensor) -> float (line 40): Mean value of difference between input values and ground truth values.

  Parameters:

  • x: type Tensor
  • y: type Tensor

  Returns: float

• threshold_test_counter(metric: Metric, threshold: float) -> Statistic[Tuple[Tensor, Tensor], str] (line 58): Create a Statistic over a stream of (x, y) pairs that counts the number of times metric(x, y) < threshold is satisfied.

  Parameters:

  • metric: Distance metric
  • threshold: Threshold to compare against

  Returns: Statistic that captures the data

• tensor_set_statistics(statistics: List[Statistic[Tuple[Any, Any], str]]) -> Statistic[Tuple[List[Any], List[Any]], str] (line 82): Apply an independent Statistic to each tensor in a stream of pairs of fixed-length lists.

  Parameters:

  • statistics: Statistic to apply to each pair of values

  Returns: Composed statistic that applies the statistics to list items

• mean(metric: Metric) -> Statistic[Tuple[List[Tensor], Tensor], float] (line 114): Create a statistic that takes input pairs (i, g) and computes the arithmetic mean of metric(i, g) over all given inputs.

  Parameters:

  • metric: Mean metric.

  Returns: Statistic that captures the data

• mean_text(metric: Metric) -> Statistic[Tuple[List[Tensor], Tensor], str] (line 136): Create a statistic that takes input pairs (i, g) and computes the arithmetic mean of metric(i, g) over all given inputs and formats the results as text message.

  Parameters:

  • metric: Mean metric.

  Returns: Statistic that captures the data