structured_data_generator package#

Submodules#

structured_data_generator.structured_data_generator module#

structured_data_generator.structured_data_generator.generate_data(fields: list, amount: int = 10, model_name: str = 'gpt-3.5-turbo', language: str = 'en', chunk_size: int = 50) → list[source]#

Structured data of elements according to the given parameters. The data can be later logged as a structured file with MLRun’s returns parameter.

Parameters

  • fields – A list of fields to randomly generate.

  • amount – The number of variants to generate.

  • model_name – The name of the model to use for conversation generation. You should choose one of GPT-4 or GPT-3.5 from the list here: https://platform.openai.com/docs/models. Default: ‘gpt-3.5-turbo’.

  • language – The language to use for the generated conversation text.

  • chunk_size – Number of samples generated at each GPT query.

Module contents#