Command Line Interface

Command-line interface for Spacylize.

This module provides the main CLI application for Spacylize, enabling users to generate, validate, visualize, split, train, and evaluate SpaCy NER datasets using LLM-powered data generation.

spacylize.cli.generate_data(llm_config_path=<typer.models.OptionInfo object>, prompt_config_path=<typer.models.OptionInfo object>, n_samples=<typer.models.OptionInfo object>, output_path=<typer.models.OptionInfo object>, task=<typer.models.OptionInfo object>)[source]

Generates training data using an LLM based on a prompt configuration.

Parameters:

  • llm_config_path (Path)

  • prompt_config_path (Path)

  • n_samples (int)

  • output_path (Path)

  • task (str)

spacylize.cli.visualize_data(input_path=<typer.models.OptionInfo object>, task=<typer.models.OptionInfo object>, n_samples=<typer.models.OptionInfo object>, port=<typer.models.OptionInfo object>)[source]

Visualizes generated data using SpaCy’s displacy.

Parameters:

  • input_path (Path)

  • task (str)

  • n_samples (int)

  • port (int)

spacylize.cli.validate_dataset(dataset=<typer.models.OptionInfo object>, output_folder=<typer.models.OptionInfo object>, task=<typer.models.OptionInfo object>)[source]

CLI entry point for validating a SpaCy dataset.

Parameters:

  • dataset (Path)

  • output_folder (Path)

  • task (str)

spacylize.cli.split_dataset(input_file=<typer.models.OptionInfo object>, train_file=<typer.models.OptionInfo object>, dev_file=<typer.models.OptionInfo object>, dev_size=<typer.models.OptionInfo object>, seed=<typer.models.OptionInfo object>)[source]

CLI entry point for splitting a SpaCy dataset.

Parameters:

  • input_file (Path)

  • train_file (Path)

  • dev_file (Path)

  • dev_size (float)

  • seed (int)

spacylize.cli.train_pipeline(train_data=<typer.models.OptionInfo object>, base_model=<typer.models.OptionInfo object>, output_model=<typer.models.OptionInfo object>, n_iter=<typer.models.OptionInfo object>, dropout=<typer.models.OptionInfo object>)[source]
Parameters:

  • train_data (Path)

  • base_model (str)

  • output_model (Path)

  • n_iter (int)

  • dropout (float)

spacylize.cli.evaluate_model(model_path=<typer.models.OptionInfo object>, eval_data=<typer.models.OptionInfo object>)[source]
Parameters:

  • model_path (Path)

  • eval_data (Path)