prevision_quantum_nn

prevision_quantum_nn module

Subpackages

• prevision_quantum_nn.applications
  • prevision_quantum_nn.applications.reinforcement_learning
    • prevision_quantum_nn.applications.reinforcement_learning.base_learner
    • prevision_quantum_nn.applications.reinforcement_learning.deep_q_learning
    • prevision_quantum_nn.applications.reinforcement_learning.policy
    • prevision_quantum_nn.applications.reinforcement_learning.q_learning
    • prevision_quantum_nn.applications.reinforcement_learning.qnn_q_learning
  • prevision_quantum_nn.applications.application
  • prevision_quantum_nn.applications.classification_application
  • prevision_quantum_nn.applications.multiclassification_application
  • prevision_quantum_nn.applications.regression_application
  • prevision_quantum_nn.applications.reinforcement_learning_application
• prevision_quantum_nn.dataset
  • prevision_quantum_nn.dataset.dataset
• prevision_quantum_nn.models
  • prevision_quantum_nn.models.pennylane_backend
    • prevision_quantum_nn.models.pennylane_backend.qnn_pennylane
    • prevision_quantum_nn.models.pennylane_backend.qnn_pennylane_cv
    • prevision_quantum_nn.models.pennylane_backend.qnn_pennylane_qubit
  • prevision_quantum_nn.models.utilities
    • prevision_quantum_nn.models.utilities.early_stopper
    • prevision_quantum_nn.models.utilities.losses
    • prevision_quantum_nn.models.utilities.to_categorical
  • prevision_quantum_nn.models.qnn
• prevision_quantum_nn.postprocessing
  • prevision_quantum_nn.postprocessing.plotter
    • prevision_quantum_nn.postprocessing.plotter.phase_space_plotter
    • prevision_quantum_nn.postprocessing.plotter.phase_space_plotter_1d
    • prevision_quantum_nn.postprocessing.plotter.phase_space_plotter_2d
  • prevision_quantum_nn.postprocessing.postprocess
• prevision_quantum_nn.preprocessing
  • prevision_quantum_nn.preprocessing.dimension_reduction
    • prevision_quantum_nn.preprocessing.dimension_reduction.pca
    • prevision_quantum_nn.preprocessing.dimension_reduction.wrapper
  • prevision_quantum_nn.preprocessing.feature_engineering
  • prevision_quantum_nn.preprocessing.polynomial_expansion
  • prevision_quantum_nn.preprocessing.preprocess
• prevision_quantum_nn.utils
  • prevision_quantum_nn.utils.get_application
  • prevision_quantum_nn.utils.get_dataset
  • prevision_quantum_nn.utils.get_jupyter_nb
  • prevision_quantum_nn.utils.get_model
  • prevision_quantum_nn.utils.get_plotter
  • prevision_quantum_nn.utils.results_parser
  • prevision_quantum_nn.utils.results_plotter

Package Contents

Functions

get_model(params)	Get a model according to parameters.
get_application(application_type, prefix=’qnn’, preprocessing_params=None, model_params=None, postprocessing_params=None, rl_learner_type=’quantum’)	Get application.
load_application(application_params, model_weights, preprocessor_file)	loads application from files
get_dataset_from_numpy(train_features, train_labels, val_features=None, val_labels=None)	get dataset from numpy
get_dataset_from_pandas(train_data_frame, targets, val_data_frame=None)	get dataset from pandas
parse_results(results_file)	Parse a result file in order to extract parameters.
plot_losses(results, prefix=’qnn’)	Plot the losses of an application.
plot_metric(results, prefix=’qnn’)	Plot the relevant metric of an application.
plot_reward(results, prefix=’qnn’)	Plot the reward of a RL application.

prevision_quantum_nn.__version__ = 1.0.1

prevision_quantum_nn.get_model(params)

Get a model according to parameters.

Parameters:params (dictionnary) – parameters of the model Returns:

QuantumNeuralNetwork

model to be constructed with these parameters

Return type:model

prevision_quantum_nn.get_application(application_type, prefix='qnn', preprocessing_params=None, model_params=None, postprocessing_params=None, rl_learner_type='quantum')

Get application.

Parameters:application_type (str) – application type can be 1. classification 2. multiclassification 3. regression 4. reinforcement_learning Returns:

Application

application according to application type

Return type:application

prevision_quantum_nn.load_application(application_params, model_weights, preprocessor_file)

loads application from files

prevision_quantum_nn.get_dataset_from_numpy(train_features, train_labels, val_features=None, val_labels=None)

get dataset from numpy

prevision_quantum_nn.get_dataset_from_pandas(train_data_frame, targets, val_data_frame=None)

get dataset from pandas

prevision_quantum_nn.parse_results(results_file)

Parse a result file in order to extract parameters.

Parameters:results_file (string) – path to a log file of an application Returns:

Pandas dataframe

dataframe containing parameters to be plotted.

Return type:results

prevision_quantum_nn.plot_losses(results, prefix='qnn')

Plot the losses of an application.

Parameters:

• results (Pandas dataframe) – dataframe containing the losses
• prefix (string) – part of the name given to the generated plot

prevision_quantum_nn.plot_metric(results, prefix='qnn')

Plot the relevant metric of an application.

Parameters:

• results (Pandas dataframe) – dataframe containing the metrics
• prefix (string) – part of the name given to the generated plot

prevision_quantum_nn.plot_reward(results, prefix='qnn')

Plot the reward of a RL application.

Parameters:

• results (Pandas dataframe) – dataframe containing the rewards
• prefix (string) – part of the name given to the generated plot