Missed TensorFlow Dev Summit? Check out the video playlist. Watch recordings

tfl.configs.CalibratedLinearConfig

View source on GitHub

Config for calibrated lattice model.

tfl.configs.CalibratedLinearConfig(
    feature_configs=None, regularizer_configs=None, use_bias=True, output_min=None,
    output_max=None, output_calibration=False, output_calibration_num_keypoints=10,
    output_initialization='quantiles'
)

Used in the notebooks

Used in the tutorials

A calibrated linear model applies piecewise-linear and categorical calibration on the input feature, followed by a linear combination and an optional output piecewise-linear calibration. When using output calibration or when output bounds are specified, the linear layer will be apply weighted averaging on calibrated inputs.

Example:

model_config = tfl.configs.CalibratedLinearConfig(
    feature_configs=[...],
)
feature_analysis_input_fn = create_input_fn(num_epochs=1, ...)
train_input_fn = create_input_fn(num_epochs=100, ...)
estimator = tfl.estimators.CannedClassifier(
    feature_columns=feature_columns,
    model_config=model_config,
    feature_analysis_input_fn=feature_analysis_input_fn)
estimator.train(input_fn=train_input_fn)

Args:

  • feature_configs: A list of tfl.configs.FeatureConfig instances that specify configurations for each feature. If a configuration is not provided for a feature, a default configuration will be used.
  • regularizer_configs: A list of tfl.configs.RegularizerConfig instances that apply global regularization.
  • use_bias: If a bias term should be used for the linear combination.
  • output_min: Lower bound constraint on the output of the model.
  • output_max: Upper bound constraint on the output of the model.
  • output_calibration: If a piecewise-linear calibration should be used on the output of the lattice.
  • output_calibration_num_keypoints: Number of keypoints to use for the output piecewise-linear calibration.
  • output_initialization: The initial values to setup for the output of the model. When using output calibration, these values are used to initliaze the output keypoints of the output piecewise-linear calibration. Otherwise the lattice parameters will be setup to form a linear function in the range of output_initialization. It can be one of:
    • String 'quantiles': Output is initliazed to label quantiles, if possible.
    • String 'uniform': Output is initliazed uniformly in label range.
    • A list of numbers: To be used for initialization of the output lattice or output calibrator.

Methods

feature_config_by_name

View source

feature_config_by_name(
    feature_name
)

Returns existing or default FeatureConfig with the given name.

regularizer_config_by_name

View source

regularizer_config_by_name(
    regularizer_name
)

Returns existing or default RegularizerConfig with the given name.