biogeme.results_processing.raw_estimation_results module

Implementation of classes containing the estimation results.

Michel Bierlaire Sun Sep 29 16:54:42 2024

class biogeme.results_processing.raw_estimation_results.RawEstimationResults(model_name, user_notes, beta_names, beta_values, lower_bounds, upper_bounds, gradient, hessian, bhhh, null_log_likelihood, initial_log_likelihood, final_log_likelihood, data_name, sample_size, number_of_observations, monte_carlo, number_of_draws, types_of_draws, number_of_excluded_data, draws_processing_time, optimization_messages, convergence, bootstrap, bootstrap_time)[source]

Bases: object

Data class containing the unprocessed estimation results. Must be simple and contain no method, in order to serialize it easily.

Parameters:

  • model_name (str)

  • user_notes (str)

  • beta_names (list[str])

  • beta_values (list[float])

  • lower_bounds (list[float])

  • upper_bounds (list[float])

  • gradient (list[float])

  • hessian (list[list[float]])

  • bhhh (list[list[float]])

  • null_log_likelihood (float)

  • initial_log_likelihood (float | None)

  • final_log_likelihood (float)

  • data_name (str)

  • sample_size (int)

  • number_of_observations (int)

  • monte_carlo (bool)

  • number_of_draws (int)

  • types_of_draws (dict[str, str])

  • number_of_excluded_data (int)

  • draws_processing_time (timedelta)

  • optimization_messages (dict[str, Any])

  • convergence (bool)

  • bootstrap (list[list[float]])

  • bootstrap_time (timedelta | None)

beta_names: list[str]
beta_values: list[float]
bhhh: list[list[float]]
bootstrap: list[list[float]]
bootstrap_time: timedelta | None
convergence: bool
data_name: str
draws_processing_time: timedelta
final_log_likelihood: float
gradient: list[float]
hessian: list[list[float]]
initial_log_likelihood: float | None
lower_bounds: list[float]
model_name: str
monte_carlo: bool
null_log_likelihood: float
number_of_draws: int
number_of_excluded_data: int
number_of_observations: int
optimization_messages: dict[str, Any]
sample_size: int
types_of_draws: dict[str, str]
upper_bounds: list[float]
user_notes: str
biogeme.results_processing.raw_estimation_results.check_for_invalid_yaml_values(data, path='root')[source]

Recursively checks for NaN or binary values in the data.

Parameters:

data (Any)

biogeme.results_processing.raw_estimation_results.contains_python_tags(yaml_string)[source]
biogeme.results_processing.raw_estimation_results.deserialize_from_yaml(filename)[source]

Restore data from a YAML file

Parameters:

filename – name of the file

Return type:

RawEstimationResults

Returns:

raw estimation results

biogeme.results_processing.raw_estimation_results.serialize_to_yaml(data, filename)[source]

Dump the data in an ASCII file

Parameters:
Return type:

None

biogeme.results_processing.raw_estimation_results.timedelta_constructor(loader, node)[source]

Construct a timedelta object from a YAML scalar representing total seconds.

Return type:

timedelta

Parameters:

  • loader (Loader)

  • node (Node)

biogeme.results_processing.raw_estimation_results.timedelta_representer(dumper, data)[source]

Represent a timedelta object as the total seconds in a YAML string.

Return type:

Node

Parameters:

  • dumper (Dumper)

  • data (timedelta)