DSPy: The framework for programming—not prompting—language models
Python
fix(MIPROv2): select between task_model and prompt_model (#8877)
Isaac Miller6374ef0bf53f
diff --git a/dspy/teleprompt/mipro_optimizer_v2.py b/dspy/teleprompt/mipro_optimizer_v2.pyindex fe741ddf8..cf76f7daf 100644--- a/dspy/teleprompt/mipro_optimizer_v2.py+++ b/dspy/teleprompt/mipro_optimizer_v2.py@@ -90,6 +90,9 @@ class MIPROv2(Teleprompter):self.seed = seedself.rng = None+ if not self.prompt_model or not self.task_model:+ raise ValueError("Either provide both prompt_model and task_model or set a default LM through dspy.configure(lm=...)")+def compile(self,student: Any,@@ -178,8 +181,9 @@ class MIPROv2(Teleprompter):provide_traceback=provide_traceback,)- # Step 1: Bootstrap few-shot examples- demo_candidates = self._bootstrap_fewshot_examples(program, trainset, seed, teacher)+ with dspy.context(lm=self.task_model):+ # Step 1: Bootstrap few-shot examples+ demo_candidates = self._bootstrap_fewshot_examples(program, trainset, seed, teacher)# Step 2: Propose instruction candidatesinstruction_candidates = self._propose_instructions(@@ -197,19 +201,20 @@ class MIPROv2(Teleprompter):if zeroshot_opt:demo_candidates = None- # Step 3: Find optimal prompt parameters- best_program = self._optimize_prompt_parameters(- program,- instruction_candidates,- demo_candidates,- evaluate,- valset,- num_trials,- minibatch,- minibatch_size,- minibatch_full_eval_steps,- seed,- )+ with dspy.context(lm=self.task_model):+ # Step 3: Find optimal prompt parameters+ best_program = self._optimize_prompt_parameters(+ program,+ instruction_candidates,+ demo_candidates,+ evaluate,+ valset,+ num_trials,+ minibatch,+ minibatch_size,+ minibatch_full_eval_steps,+ seed,+ )return best_program