synthetic.nemotron

Module Contents

Classes

NemotronFormatter
NemotronGenerator

API

class synthetic.nemotron.NemotronFormatter

Bases: nemo_curator.services.conversation_formatter.ConversationFormatter

PROMPT_PREFIX = <Multiline-String>

static format_conversation(conv: list[dict]) → str

class synthetic.nemotron.NemotronGenerator(

llm_client: nemo_curator.services.model_client.LLMClient,

)

Initialization

classify_math_entity(

entity: str,

model: str,

prompt_template: str = DEFAULT_MATH_CLASSIFICATION_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

classify_python_entity(

entity: str,

model: str,

prompt_template: str = DEFAULT_PYTHON_CLASSIFICATION_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

convert_response_to_yaml_list(

llm_response: str,

model: str,

prompt_template: str = DEFAULT_YAML_CONVERSION_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

generate_closed_qa_instructions(

document: str,

n_openlines: str | int,

model: str,

prompt_template: str = DEFAULT_CLOSED_QA_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

generate_dialogue(

openline: str,

user_model: str,

assistant_model: str,

n_user_turns: int = 3,

prompt_template: str = DIALOGUE_NORMAL_USER_TURN_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

user_model_kwargs: dict | None = None,

assistant_model_kwargs: dict | None = None,

) → list[dict]

generate_macro_topics(

n_macro_topics: int | str,

model: str,

prompt_template: str = DEFAULT_MACRO_TOPICS_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

generate_math_macro_topics(

n_macro_topics: int | str,

school_level: str,

model: str,

prompt_template: str = DEFAULT_MATH_MACRO_TOPICS_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

generate_math_problem(

topic: str,

n_openlines: str | int,

model: str,

prompt_template: str = MATH_PROBLEM_GENERAL_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

generate_math_subtopics(

macro_topic: str,

n_subtopics: int | str,

model: str,

prompt_template: str = DEFAULT_MATH_SUBTOPICS_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

generate_open_qa_from_topic(

topic: str,

n_openlines: str | int,

model: str,

prompt_template: str = DEFAULT_OPEN_QA_FROM_TOPICS_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

generate_python_macro_topics(

n_macro_topics: int | str,

model: str,

prompt_template: str = DEFAULT_PYTHON_MACRO_TOPICS_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

generate_python_problem(

topic: str,

n_openlines: str | int,

model: str,

language: str = 'Python',

prompt_template: str = PYTHON_PROBLEM_BEGINNER_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

generate_python_subtopics(

macro_topic: str,

n_subtopics: int | str,

model: str,

prompt_template: str = DEFAULT_PYTHON_SUBTOPICS_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

generate_subtopics(

macro_topic: str,

n_subtopics: int | str,

model: str,

prompt_template: str = DEFAULT_SUBTOPICS_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

generate_two_turn_prompt(

openline: str,

user_model: str,

assistant_model: str,

prompt_template: str = DIALOGUE_NORMAL_USER_TURN_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

user_model_kwargs: dict | None = None,

assistant_model_kwargs: dict | None = None,

) → list[dict]

generate_writing_tasks(

topic: str,

text_material_type: str,

n_openlines: str | int,

model: str,

prompt_template: str = DEFAULT_WRITING_TASK_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

revise_open_qa(

openline: str,

n_revisions: str | int,

model: str,

prompt_template: str = DEFAULT_REVISE_OPEN_QA_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

revise_writing_tasks(

openline: str,

n_revisions: str | int,

model: str,

prompt_template: str = DEFAULT_REVISE_WRITING_TASK_PROMPT_TEMPLATE,

prompt_kwargs: dict | None = None,

model_kwargs: dict | None = None,

) → list[str]

run_closed_qa_pipeline(

documents: list[str],

n_openlines: str | int,

model: str,

closed_qa_prompt_template: str = DEFAULT_CLOSED_QA_PROMPT_TEMPLATE,

yaml_conversion_prompt_template: str = DEFAULT_YAML_CONVERSION_PROMPT_TEMPLATE,

base_model_kwargs: dict | None = None,

conversion_model_kwargs: dict | None = None,

ignore_conversion_failure: bool = False,

) → list[tuple[int, str]]

run_math_pipeline(

n_macro_topics: str | int,

school_level: str,

n_subtopics: str | int,

n_openlines: str | int,

model: str,

macro_topic_prompt_template: str = DEFAULT_MATH_MACRO_TOPICS_PROMPT_TEMPLATE,

subtopic_prompt_template: str = DEFAULT_MATH_SUBTOPICS_PROMPT_TEMPLATE,

math_problem_prompt_template: str = MATH_PROBLEM_GENERAL_PROMPT_TEMPLATE,

yaml_conversion_prompt_template: str = DEFAULT_YAML_CONVERSION_PROMPT_TEMPLATE,

base_model_kwargs: dict | None = None,

conversion_model_kwargs: dict | None = None,

additional_macro_topics: list[str] | None = None,

additional_subtopics: list[str] | None = None,

ignore_conversion_failure: bool = False,

combine_topics: bool = True,

) → list[str]

run_open_qa_pipeline(

n_macro_topics: str | int,

n_subtopics: str | int,

n_openlines: str | int,

n_revisions: str | int,

model: str,

macro_topic_prompt_template: str = DEFAULT_MACRO_TOPICS_PROMPT_TEMPLATE,

subtopic_prompt_template: str = DEFAULT_SUBTOPICS_PROMPT_TEMPLATE,

open_qa_from_topics_prompt_template: str = DEFAULT_OPEN_QA_FROM_TOPICS_PROMPT_TEMPLATE,

revise_open_qa_prompt_template: str = DEFAULT_REVISE_OPEN_QA_PROMPT_TEMPLATE,

yaml_conversion_prompt_template: str = DEFAULT_YAML_CONVERSION_PROMPT_TEMPLATE,

base_model_kwargs: dict | None = None,

conversion_model_kwargs: dict | None = None,

additional_macro_topics: list[str] | None = None,

additional_subtopics: list[str] | None = None,

ignore_conversion_failure: bool = False,

combine_topics: bool = True,

) → list[str]

run_python_pipeline(

n_macro_topics: str | int,

n_subtopics: str | int,

n_openlines: str | int,

model: str,

macro_topic_prompt_template: str = DEFAULT_PYTHON_MACRO_TOPICS_PROMPT_TEMPLATE,

subtopic_prompt_template: str = DEFAULT_PYTHON_SUBTOPICS_PROMPT_TEMPLATE,

python_problem_prompt_template: str = PYTHON_PROBLEM_BEGINNER_PROMPT_TEMPLATE,

yaml_conversion_prompt_template: str = DEFAULT_YAML_CONVERSION_PROMPT_TEMPLATE,

base_model_kwargs: dict | None = None,

conversion_model_kwargs: dict | None = None,

additional_macro_topics: list[str] | None = None,

additional_subtopics: list[str] | None = None,

ignore_conversion_failure: bool = False,

combine_topics: bool = True,

) → list[str]

run_writing_pipeline(

topics: list[str],

text_material_types: list[str],

n_openlines: str | int,

n_revisions: str | int,

model: str,

writing_task_prompt_template: str = DEFAULT_WRITING_TASK_PROMPT_TEMPLATE,

revise_writing_task_prompt_template: str = DEFAULT_REVISE_WRITING_TASK_PROMPT_TEMPLATE,

yaml_conversion_prompt_template: str = DEFAULT_YAML_CONVERSION_PROMPT_TEMPLATE,

base_model_kwargs: dict | None = None,

conversion_model_kwargs: dict | None = None,

ignore_conversion_failure: bool = False,

) → list[str]