Inversion

class kgpy.ctis.inversions.cnn.Inversion(instrument, instrument_inverse, history, average_input, average_output, width_input, width_output, num_divisions_fov)

Bases: Pickleable, AbstractInversion

Parameters:
__init__(instrument, instrument_inverse, history, average_input, average_output, width_input, width_output, num_divisions_fov)
Parameters:
Return type:

None

Attributes

instrument_inverse

history

average_input

average_output

width_input

width_output

num_divisions_fov

instrument

Methods

__init__(instrument, instrument_inverse, ...)

from_pickle([path])

instrument_inverse_initial(input_shape[, ...])

rtype:

keras.Sequential

to_pickle(path)

train(instrument, scene_training, ...[, ...])

rtype:

InversionT

Inheritance Diagram

digraph inheritance25aaaf9943 { bgcolor=transparent; rankdir=TB; size="8.0, 12.0"; "abc.ABC" [URL="https://docs.python.org/3/library/abc.html#abc.ABC",fillcolor=white,fontname="Vera Sans, DejaVu Sans, Liberation Sans, Arial, Helvetica, sans",fontsize=10,height=0.25,shape=box,style="setlinewidth(0.5),filled",target="_top",tooltip="Helper class that provides a standard way to create an ABC using"]; "kgpy.ctis.inversions.abstractions.AbstractInversion" [URL="kgpy.ctis.inversions.abstractions.AbstractInversion.html#kgpy.ctis.inversions.abstractions.AbstractInversion",fillcolor=white,fontname="Vera Sans, DejaVu Sans, Liberation Sans, Arial, Helvetica, sans",fontsize=10,height=0.25,shape=box,style="setlinewidth(0.5),filled",target="_top",tooltip="AbstractInversion(instrument: kgpy.ctis.instruments.AbstractInstrument)"]; "abc.ABC" -> "kgpy.ctis.inversions.abstractions.AbstractInversion" [arrowsize=0.5,style="setlinewidth(0.5)"]; "kgpy.ctis.inversions.cnn.Inversion" [URL="kgpy.ctis.inversions.cnn.Inversion.html#kgpy.ctis.inversions.cnn.Inversion",fillcolor=white,fontname="Vera Sans, DejaVu Sans, Liberation Sans, Arial, Helvetica, sans",fontsize=10,height=0.25,shape=box,style="setlinewidth(0.5),filled",target="_top",tooltip="Inversion(instrument: kgpy.ctis.instruments.AbstractInstrument, instrument_inverse: 'keras.Sequential', history: 'TrainingHistory', average_input: 'kgpy.labeled.AbstractArray', average_output: 'kgpy.labeled.AbstractArray', width_input: 'kgpy.labeled.AbstractArray', width_output: 'kgpy.labeled.AbstractArray', num_divisions_fov: 'kgpy.vectors.Cartesian2D')"]; "kgpy.mixin.Pickleable" -> "kgpy.ctis.inversions.cnn.Inversion" [arrowsize=0.5,style="setlinewidth(0.5)"]; "kgpy.ctis.inversions.abstractions.AbstractInversion" -> "kgpy.ctis.inversions.cnn.Inversion" [arrowsize=0.5,style="setlinewidth(0.5)"]; "kgpy.mixin.Pickleable" [URL="kgpy.mixin.Pickleable.html#kgpy.mixin.Pickleable",fillcolor=white,fontname="Vera Sans, DejaVu Sans, Liberation Sans, Arial, Helvetica, sans",fontsize=10,height=0.25,shape=box,style="setlinewidth(0.5),filled",target="_top",tooltip="Class for adding 'to_pickle' and 'from_pickle' methods for objects with long creation times."]; "abc.ABC" -> "kgpy.mixin.Pickleable" [arrowsize=0.5,style="setlinewidth(0.5)"]; }
classmethod from_pickle(path=None)
Parameters:

path (Path | None) –

static instrument_inverse_initial(input_shape, n_filters=32, kernel_size=7, growth_factor=2, alpha=0.1, dropout_rate=0.01)
Return type:

keras.Sequential

Parameters:

  • input_shape (tuple[int | None, int | None, int | None, int | None]) –

  • n_filters (int) –

  • kernel_size (int) –

  • growth_factor (int) –

  • alpha (float) –

  • dropout_rate (float) –

to_pickle(path)
Parameters:

path (Path | None) –

classmethod train(instrument, scene_training, scene_validation, deprojections_training, deprojections_validation, instrument_inverse=None, num_divisions_fov=None, epochs=1000)
Return type:

InversionT

Parameters:
average_input: kgpy.labeled.AbstractArray
average_output: kgpy.labeled.AbstractArray
history: kgpy.ctis.inversions.cnn.TrainingHistory
instrument: kgpy.ctis.instruments.AbstractInstrument
instrument_inverse: keras.src.engine.sequential.Sequential
num_divisions_fov: kgpy.vectors.Cartesian2D
width_input: kgpy.labeled.AbstractArray
width_output: kgpy.labeled.AbstractArray