turicreate.toolkits.distances.transformed_dot_product — Turi Create API 6.4.1 documentation (original) (raw)

• Data
  • Data Structures
  • Data Types
  • Utilities
• Modelling
  • Applications
  • Essentials
  • Utilities
• Visualization
• Configuration
  • turicreate.config.get_client_log_location
  • turicreate.config.get_environment_config
  • turicreate.config.get_num_gpus
  • turicreate.config.get_runtime_config
  • turicreate.config.set_log_level
  • turicreate.config.set_num_gpus
  • turicreate.config.set_runtime_config

Turi Create API

• Docs »
• Modelling »
• distances »
• turicreate.toolkits.distances.transformed_dot_product

turicreate.toolkits.distances. transformed_dot_product(x, y)¶

Compute the “transformed_dot_product” distance between two dictionaries or two lists of equal length. This is a way to transform the dot product of the two inputs—a similarity measure—into a distance measure. Suppose x andy each contain \(d\) variables:

\[D(x, y) = \log\{1 + \exp\{-\sum_i^d x_i y_i\}\}\]

Warning

The ‘dot_product’ distance is deprecated and will be removed in future versions of Turi Create. Please use ‘transformed_dot_product’ distance instead, although note that this is more than a name change; it is a different transformation of the dot product of two vectors. Please see the distances module documentation for more details.

Notes

• If the input vectors are in dictionary form, keys missing in one of the two dictionaries are assumed to have value 0.
• Transformed dot product distance is not a metric because the distance from a point to itself is not 0. This means the ball tree cannot be used to compute nearest neighbors based on this distance.

Examples

tc.distances.transformed_dot_product([1, 2, 3], [4, 5, 6]) 0.03125 ... tc.distances.transformed_dot_product({'a': 2, 'c': 4}, {'b': 3, 'c': 12}) 0.020833333333333332