opto.optimizers.optoprime.OptoPrime#

class OptoPrime[source]#

Attributes

default_objective
default_prompt_symbols
example_problem_template
example_prompt
final_prompt
output_format_prompt
propagator	Return a Propagator object that can be used to propagate feedback in backward.
representation_prompt
trace_graph	Aggregate the graphs of all the parameters.
user_prompt_template

Methods

backward(node, *args, **kwargs)	Propagate the feedback backward.
call_llm(system_prompt, user_prompt[, ...])	Call the LLM with a prompt and return the response.
construct_prompt(summary[, mask])	Construct the system and user prompt.
construct_update_dict(suggestion)	Convert the suggestion in text into the right data type.
default_propagator()	Return the default Propagator object of the optimizer.
extract_llm_suggestion(response)	Extract the suggestion from the response.
problem_instance(summary[, mask])
propose(*args, **kwargs)	Propose the new data of the parameters based on the feedback.
replace_symbols(text, symbols)
repr_node_constraint(node_dict)
repr_node_value(node_dict)
step(*args, **kwargs)	Update the parameters based on the feedback.
summarize()
update(update_dict)	Update the trainable parameters given a dictionary of new data.
zero_feedback()	Reset the feedback.

__init__(parameters: List[ParameterNode], llm: AutoGenLLM | None = None, *args, propagator: Propagator | None = None, objective: None | str = None, ignore_extraction_error: bool = True, include_example=False, memory_size=0, max_tokens=4096, log=True, prompt_symbols=None, filter_dict: Dict | None = None, **kwargs)[source]#

__new__(**kwargs)#