tommy.controller.graph_controller
Classes
- class tommy.controller.graph_controller.GraphController[source]
Bases:
objectThe central interface for extracting results from topic modelling results and creating visualizations. Contains an event for when the plots are changed and a publisher for when the topics have changed.
- NX_EXPORTS: list[NxExporterOnData | NxExporter] = [<tommy.controller.visualizations.document_topic_nx_exporter.DocumentTopicNxExporter object>, <tommy.controller.visualizations.word_topic_nx_exporter.WordTopicNxExporter object>]
- VISUALIZATIONS: list[AbstractVisualization] = [<tommy.controller.visualizations.document_word_count_creator.DocumentWordCountCreator object>, <tommy.controller.visualizations.k_value_creator.KValueCreator object>, <tommy.controller.visualizations.documents_over_time_creator.DocumentsOverTimeCreator object>, <tommy.controller.visualizations.sum_topics_in_documents.SumTopicsInDocuments object>, <tommy.controller.visualizations.correlation_matrix_creator.CorrelationMatrixCreator object>, <tommy.controller.visualizations.word_topic_network_creator.WordTopicNetworkCreator object>, <tommy.controller.visualizations.document_topic_network_summary_creator.DocumentTopicNetworkSummaryCreator object>, <tommy.controller.visualizations.word_cloud_creator.WordCloudCreator object>, <tommy.controller.visualizations.top_words_bar_plot_creator.TopWordsBarPlotCreator object>, <tommy.controller.visualizations.documents_over_time_per_topic_creator.DocumentsOverTimePerTopicCreator object>, <tommy.controller.visualizations.welcome_screen.WelcomeScreen object>]
- get_all_nx_exports() list[NxExport][source]
Get all the networkx graphs for the possible visualization for the current run :return: A list of nx.graph objects of all possible visualizations
- get_all_visualizations(ignore_cache: bool = False) list[MatplotLibExport][source]
Get all the possible visualization for the current run :param ignore_cache: Whether to ignore the cache and always create a
new figure, defaults to False
- Returns:
A list of matplotlib Figures of all possible visualizations
- get_model_type() str[source]
Get the model type in the topic modelling results :return: the model type in the topic modelling results :raises RuntimeError: if the topic runner has not finished running yet.
- get_number_of_topics() int[source]
Get the number of topics in the topic modelling results :return: the number of topics in the topic modelling results :raises RuntimeError: if the topic runner has not finished running yet.
- get_topic_name(topic_index: int) str[source]
Get the name of the topic with the given index :param topic_index: The index of the topic :return: The name of the topic
- get_topic_with_scores(topic_id, n_words) TopicWithScores[source]
Return a topic object containing top n terms and their corresponding score for the topic identified by the topic_id.
- Parameters:
topic_id – the index of the requested topic
n_words – number of terms in the resulting topic object
- Returns:
topic object containing top n terms and their corresponding scores
- get_visualization(vis_index: int, override_topic: int | None = None, ignore_cache: bool = False) tuple[Figure, str][source]
Returns the visualization corresponding to the given index in the list of all visualizations. :param vis_index: Index of the visualization to be requested :param override_topic: A topic index used to override the selected
topic, default to None, which doesn’t override the selected topic
- Parameters:
ignore_cache – Whether to ignore the cache and always create a new figure, defaults to False
- Returns:
matplotlib figure of visualization corresponding to the index
and the type of the visualization :raises IndexError: if the index is negative or bigger than the number
of visualizations or if the visualization corresponding to that index is not possible in the current topic model.
- property has_topic_runner: bool
- on_new_topic_runner(topic_runner: TopicRunner) None[source]
Signal the graph-controller that a topic runner has finished training and is ready to provide results. Notify the subscribes of the plots and topics :param topic_runner: The newly trained topic runner object :return: None
- on_topic_runner_switched(topic_runner: TopicRunner) None[source]
Signal the graph-controller that a topic runner has been switched :param topic_runner: The newly trained topic runner object :return: None
- property possible_plots_changed_event: EventHandler[list[PossibleVisualization]]
Get event that triggers when the list of possible plots changes.
- property refresh_name_event: EventHandler[None]
Get the event that triggers when the config name changes.
- property refresh_plots_event: EventHandler[None]
Get the event that triggers when the content of any plots changes.
- set_controller_refs(topic_modelling_controller: TopicModellingController, corpus_controller: CorpusController, project_settings_controller: ProjectSettingsController) None[source]
Set reference to the TM controller corpus controller and add self to model trained event
- set_model_refs(topic_name_model: TopicNameModel) None[source]
Set the reference to the topic name model :param topic_name_model: The topic name model :return: None
- set_selected_topic(topic_index: int | None) None[source]
Set the currently selected topic to the given index or None if no topic is to be selected.
- Parameters:
topic_index – the index of the topic to select
- Returns:
None
- set_topic_name(topic_index: int, name: str) None[source]
Set the name of the topic with the given index :param topic_index: The index of the topic :param name: The new name of the topic :return: None
- property topics_changed_event: EventHandler[None]
Get the eventhandler that triggers when the topics are changed.