BERTopicPredict
Python template · sinapsis_bertopic.templates.bertopic_predict
Overview
BERTopicPredict template for topic prediction using BERTopic models.
This class provides functionality to load a pre-trained BERTopic model and generate topic predictions for document collections. It handles model loading, prediction generation, and integration with the data container pipeline.
Attributes: AttributesBaseModel: Configuration model for BERTopicPredict template. - model_path (str): Path to the saved BERTopic model file. It can point to a local saved model or a model stored in a huggingface hub repo. - visualize_predictions (bool): Whether to generate visualizations for predictions. Defaults to False. - historical_data_path (str | None): Path to historical data required for visualization. Must be provided if visualize_predictions is True. Defaults to None. - root_dir (str): Root directory for storing prediction outputs. Defaults to SINAPSIS_CACHE_DIR. - prediction_viz_path (str): Relative path for the prediction visualization file. Defaults to "prediction_viz.html". - save_topic_visualization_path (str): Relative path for the topic visualization file. Defaults to "topics_visualization.html".
Capabilities
Capabilities: not setInstallation
Install the published package using pip or uv.
pip install sinapsis-bertopic==1.0.0uv pip install sinapsis-bertopic==1.0.0uv add sinapsis-bertopic==1.0.0Configuration
Effective fields from the inspector schema, including nested Pydantic model fields.
Optional
Default: null
Root directory for storing prediction outputs. Defaults to SINAPSIS_CACHE_DIR.
Optional
Default: "/home/ariel/.cache/sinapsis"
Path to the trained BERTopic model file.
Required
Default: null
Optional
Default: true
Optional
Default: null
Optional
Default: null
Optional
Default: null
Optional
Default: "png"
Optional
Default: null
Optional
Default: true
Relative path for the prediction visualization file. Defaults to "prediction_viz.html". Methods: validate_attributes() -> None: Validates that historical_data_path is provided when visualize_predictions is enabled.… Read full field description
Relative path for the prediction visualization file. Defaults to "prediction_viz.html". Methods: validate_attributes() -> None: Validates that historical_data_path is provided when visualize_predictions is enabled. Raises ValueError if validation fails. full_prediction_viz_path (property): Computed field that combines root_dir and prediction_viz_path into an absolute file path.
Optional
Default: "prediction_viz.html"
Optional
Default: "sentence-transformers/all-MiniLM-L6-v2"
Path to historical data required for visualization. Must be provided if visualize_predictions is True. Defaults to None.
Optional
Default: null
Whether to generate visualizations for predictions. Defaults to False.
Optional
Default: false
Optional
Default: null
Optional
Default: "<b>Intertopic Distance Map</b>"
Optional
Default: null
Optional
Default: false
Optional
Default: 650
Optional
Default: null
Optional
Default: false
Optional
Default: 650
Optional
Default: true
Optional
Default: "topics_visualization.html"
Configuration example
Generated from documented defaults. Replace any placeholders and complete required nested settings before running.
templates:
- template_name: "BERTopicPredict"
class_name: "BERTopicPredict"
attributes:
root_dir: "/home/ariel/.cache/sinapsis"
model_path: "<required>"
visualize_topics: true
image_export_params:
scale: null
width: null
format: "png"
height: null
validate_figure: true
prediction_viz_path: "prediction_viz.html"
sentence_model_name: "sentence-transformers/all-MiniLM-L6-v2"
historical_data_path: null
visualize_predictions: false
visualize_topics_params:
title: "<b>Intertopic Distance Map</b>"
topics: null
use_ctfidf: false
figure_width: 650
top_n_topics: null
custom_labels: false
figure_height: 650
export_visualization_to_image: true
save_topic_visualization_path: "topics_visualization.html"
Inheritance
sinapsis_bertopic.templates.bertopic_base.BERTopicBaseEnvironment variables
Environment definitions belong to their canonical Python modules.
No directly related environment variables listed.