DispReconstructor

class ctapipe.reco.sklearn.DispReconstructor(**kwargs: Any)

Bases: Reconstructor

Predict absolute value and sign for disp origin reconstruction and convert to altitude and azimuth prediction for each telescope.

Attributes Summary

features	Features to use for both models.
instrument_table
load_path	If given, load serialized model from this path.
log_target	If True, the norm(disp) model is trained to predict ln(norm(disp)) and the output is prefix_parameter = sign_prediction * exp(norm_prediction).
norm_cls	Which scikit-learn regression model to use.
norm_config	kwargs for the sklearn regressor.
prefix	Prefix for the output of this model.
sign_cls	Which scikit-learn classification model to use.
sign_config	kwargs for the sklearn classifier.
stereo_combiner_cls	Which stereo combination method to use.
target

Methods Summary

__call__(event)	Event-wise prediction for the EventSource-Loop.
fit(key, table)	Create and fit new models for key using the data in table.
predict_table(key, table)	Predict on a table of events.
read(path, **kwargs)	Read a joblib-pickled reconstructor from path
write(path[, overwrite])

Attributes Documentation

features

Features to use for both models.

instrument_table

load_path

If given, load serialized model from this path.

log_target

If True, the norm(disp) model is trained to predict ln(norm(disp)) and the output is prefix_parameter = sign_prediction * exp(norm_prediction).

norm_cls

Which scikit-learn regression model to use.

norm_config

kwargs for the sklearn regressor.

prefix

Prefix for the output of this model. If None, disp is used.

sign_cls

Which scikit-learn classification model to use.

sign_config

kwargs for the sklearn classifier.

stereo_combiner_cls

Which stereo combination method to use.. Possible values: []

target = 'true_disp'

Methods Documentation

__call__(event: ArrayEventContainer) → None

Event-wise prediction for the EventSource-Loop.

Fills the event.dl2.tel[tel_id].disp[prefix] container and event.dl2.tel[tel_id].geometry[prefix] container.

Parameters:

event: ArrayEventContainer

fit(key, table)

Create and fit new models for key using the data in table.

predict_table(key, table: Table) → dict[ReconstructionProperty, Table]

Predict on a table of events.

Parameters:

tableTable

Table of features

Returns:

disp_tableTable

Table with disp predictions, matches the corresponding container definition

altaz_tableTable

Table with resulting predictions of horizontal coordinates

classmethod read(path, **kwargs)

Read a joblib-pickled reconstructor from path

Parameters:

pathstr or pathlib.Path

Path to a Reconstructor instance pickled using joblib

parentNone or Component or Tool

Attach a new parent to the loaded class, this will properly

subarraySubarrayDescription

Attach a new subarray to the loaded reconstructor A warning will be raised if the telescope types of the subarray stored in the pickled class do not match with the provided subarray.

**kwargs are set on the loaded instance

Returns:

Reconstructor instance loaded from file

write(path, overwrite=False)