pickling#
This file contains the methods to be used to save predictors and classifiers to files that can be transferred back to Blueback Investigator
Functions
load_classifier(filename)Loads a function from 'filename'
load_predictor(filename)Loads a function from 'filename'
save_classifier(filename, func, inputs, output)Saves a classifier function to 'filename'
save_predictor(filename, func, inputs, output)Saves a predictor function to 'filename'
- cegalprizm.investigator.pickling.load_classifier(filename: str)#
Loads a function from ‘filename’
- Parameters
filename (str) – The filename
- cegalprizm.investigator.pickling.load_predictor(filename: str)#
Loads a function from ‘filename’
- Returns
file
- Return type
cloudpickle
- cegalprizm.investigator.pickling.save_classifier(filename: str, func: Callable, inputs: Sequence[tuple], output: tuple)#
Saves a classifier function to ‘filename’
- Parameters
filename (str) – the name of the file to be created
func (function) – the function that should be invoked to run the classifier
inputs (list) – a list of tuples each containing a continuous dimension name and a unit string (see Investigation.available_units property)
output (tuple) – a tuple of discrete dimension name, list of discrete tags (see Investigation.discrete_dimension_tags property)
- Output:
json: A JSON file containing all the required information to define the classifier and the pickled classifier function
- cegalprizm.investigator.pickling.save_predictor(filename: str, func: Callable, inputs: Sequence[tuple], output: tuple)#
Saves a predictor function to ‘filename’
- Parameters
filename (str) – the name of the file to be created
func (function) – the name of the function that should be invoked to run the predictor
inputs (list) – a list of tuples each containing a continuous dimension name and a unit string (see Investigation.available_units property)
output (tuple) – a tuple of continuous dimension name, a unit string (see Investigation.available_units property)
- Outputs:
json: A JSON file containing all the required information to define the predictor and the pickled predictor function