DSPy: The framework for programming—not prompting—language models
Python
Adds Improved Error Handling, Documentation, and Tests For The BestOfN Module (#7964)
Cezar Cocu95947c62dfb9
diff --git a/docs/docs/api/modules/BestOfN.md b/docs/docs/api/modules/BestOfN.mdnew file mode 100644index 000000000..bb7ca3b5b--- /dev/null+++ b/docs/docs/api/modules/BestOfN.md@@ -0,0 +1,32 @@+# dspy.BestOfN++::: dspy.BestOfN+ handler: python+ options:+ members:+ - __call__+ - batch+ - deepcopy+ - dump_state+ - forward+ - get_lm+ - load+ - load_state+ - map_named_predictors+ - named_parameters+ - named_predictors+ - named_sub_modules+ - parameters+ - predictors+ - reset_copy+ - save+ - set_lm+ show_source: true+ show_undocumented_members: true+ show_root_heading: true+ show_inherited_members: true+ heading_level: 2+ docstring_style: google+ show_root_full_path: true+ show_object_full_path: false+ separate_signature: falsediff --git a/docs/docs/api/modules/Refine.md b/docs/docs/api/modules/Refine.mdnew file mode 100644index 000000000..7e27a3b0a--- /dev/null+++ b/docs/docs/api/modules/Refine.md@@ -0,0 +1,32 @@+# dspy.Refine++::: dspy.Refine+ handler: python+ options:+ members:+ - __call__+ - batch+ - deepcopy+ - dump_state+ - forward+ - get_lm+ - load+ - load_state+ - map_named_predictors+ - named_parameters+ - named_predictors+ - named_sub_modules+ - parameters+ - predictors+ - reset_copy+ - save+ - set_lm+ show_source: true+ show_undocumented_members: true+ show_root_heading: true+ show_inherited_members: true+ heading_level: 2+ docstring_style: google+ show_root_full_path: true+ show_object_full_path: false+ separate_signature: falsediff --git a/docs/docs/deep-dive/assertions.md b/docs/docs/deep-dive/assertions.mddeleted file mode 100644index 858d03780..000000000--- a/docs/docs/deep-dive/assertions.md+++ /dev/null@@ -1,268 +0,0 @@-----sidebar_position: 7------# DSPy Assertions--!!! warning "This page is outdated and may not be fully accurate in DSPy 2.5"---Language models (LMs) have transformed how we interact with machine learning, offering vast capabilities in natural language understanding and generation. However, ensuring these models adhere to domain-specific constraints remains a challenge. Despite the growth of techniques like fine-tuning or “prompt engineering”, these approaches are extremely tedious and rely on heavy, manual hand-waving to guide the LMs in adhering to specific constraints. Even DSPy's modularity of programming prompting pipelines lacks mechanisms to effectively and automatically enforce these constraints.--To address this, we introduce DSPy Assertions, a feature within the DSPy framework designed to automate the enforcement of computational constraints on LMs. DSPy Assertions empower developers to guide LMs towards desired outcomes with minimal manual intervention, enhancing the reliability, predictability, and correctness of LM outputs.--## dspy.Assert and dspy.Suggest API--We introduce two primary constructs within DSPy Assertions:--- **`dspy.Assert`**:- - **Parameters**:- - `constraint (bool)`: Outcome of Python-defined boolean validation check.- - `msg (Optional[str])`: User-defined error message providing feedback or correction guidance.- - `backtrack (Optional[module])`: Specifies target module for retry attempts upon constraint failure. The default backtracking module is the last module before the assertion.- - **Behavior**: Initiates retry upon failure, dynamically adjusting the pipeline's execution. If failures persist, it halts execution and raises a `dspy.AssertionError`.--- **`dspy.Suggest`**:- - **Parameters**: Similar to `dspy.Assert`.- - **Behavior**: Encourages self-refinement through retries without enforcing hard stops. Logs failures after maximum backtracking attempts and continues execution.--- **dspy.Assert vs. Python Assertions**: Unlike conventional Python `assert` statements that terminate the program upon failure, `dspy.Assert` conducts a sophisticated retry mechanism, allowing the pipeline to adjust.--Specifically, when a constraint is not met:--- Backtracking Mechanism: An under-the-hood backtracking is initiated, offering the model a chance to self-refine and proceed, which is done through-- Dynamic Signature Modification: internally modifying your DSPy program’s Signature by adding the following fields:- - Past Output: your model's past output that did not pass the validation_fn- - Instruction: your user-defined feedback message on what went wrong and what possibly to fix--If the error continues past the `max_backtracking_attempts`, then `dspy.Assert` will halt the pipeline execution, altering you with an `dspy.AssertionError`. This ensures your program doesn't continue executing with “bad” LM behavior and immediately highlights sample failure outputs for user assessment.--- **dspy.Suggest vs. dspy.Assert**: `dspy.Suggest` on the other hand offers a softer approach. It maintains the same retry backtracking as `dspy.Assert` but instead serves as a gentle nudger. If the model outputs cannot pass the model constraints after the `max_backtracking_attempts`, `dspy.Suggest` will log the persistent failure and continue execution of the program on the rest of the data. This ensures the LM pipeline works in a "best-effort" manner without halting execution.--- **`dspy.Suggest`** are best utilized as "helpers" during the evaluation phase, offering guidance and potential corrections without halting the pipeline.-- **`dspy.Assert`** are recommended during the development stage as "checkers" to ensure the LM behaves as expected, providing a robust mechanism for identifying and addressing errors early in the development cycle.---## Use Case: Including Assertions in DSPy Programs--We start with using an example of a multi-hop QA SimplifiedBaleen pipeline as defined in the intro walkthrough.--```python-class SimplifiedBaleen(dspy.Module):- def __init__(self, passages_per_hop=2, max_hops=2):- super().__init__()-- self.generate_query = [dspy.ChainOfThought(GenerateSearchQuery) for _ in range(max_hops)]- self.retrieve = dspy.Retrieve(k=passages_per_hop)- self.generate_answer = dspy.ChainOfThought(GenerateAnswer)- self.max_hops = max_hops-- def forward(self, question):- context = []- prev_queries = [question]-- for hop in range(self.max_hops):- query = self.generate_query[hop](context=context, question=question).query- prev_queries.append(query)- passages = self.retrieve(query).passages- context = deduplicate(context + passages)-- pred = self.generate_answer(context=context, question=question)- pred = dspy.Prediction(context=context, answer=pred.answer)- return pred--baleen = SimplifiedBaleen()--baleen(question = "Which award did Gary Zukav's first book receive?")-```--To include DSPy Assertions, we simply define our validation functions and declare our assertions following the respective model generation.--For this use case, suppose we want to impose the following constraints:- 1. Length - each query should be less than 100 characters- 2. Uniqueness - each generated query should differ from previously-generated queries.--We can define these validation checks as boolean functions:--```python-#simplistic boolean check for query length-len(query) <= 100--#Python function for validating distinct queries-def validate_query_distinction_local(previous_queries, query):- """check if query is distinct from previous queries"""- if previous_queries == []:- return True- if dspy.evaluate.answer_exact_match_str(query, previous_queries, frac=0.8):- return False- return True-```--We can declare these validation checks through `dspy.Suggest` statements (as we want to test the program in a best-effort demonstration). We want to keep these after the query generation `query = self.generate_query[hop](context=context, question=question).query`.--```python-dspy.Suggest(- len(query) <= 100,- "Query should be short and less than 100 characters",- target_module=self.generate_query-)--dspy.Suggest(- validate_query_distinction_local(prev_queries, query),- "Query should be distinct from: "- + "; ".join(f"{i+1}) {q}" for i, q in enumerate(prev_queries)),- target_module=self.generate_query-)-```--It is recommended to define a program with assertions separately than your original program if you are doing comparative evaluation for the effect of assertions. If not, feel free to set Assertions away!--Let's take a look at how the SimplifiedBaleen program will look with Assertions included:--```python-class SimplifiedBaleenAssertions(dspy.Module):- def __init__(self, passages_per_hop=2, max_hops=2):- super().__init__()- self.generate_query = [dspy.ChainOfThought(GenerateSearchQuery) for _ in range(max_hops)]- self.retrieve = dspy.Retrieve(k=passages_per_hop)- self.generate_answer = dspy.ChainOfThought(GenerateAnswer)- self.max_hops = max_hops-- def forward(self, question):- context = []- prev_queries = [question]-- for hop in range(self.max_hops):- query = self.generate_query[hop](context=context, question=question).query-- dspy.Suggest(- len(query) <= 100,- "Query should be short and less than 100 characters",- target_module=self.generate_query- )-- dspy.Suggest(- validate_query_distinction_local(prev_queries, query),- "Query should be distinct from: "- + "; ".join(f"{i+1}) {q}" for i, q in enumerate(prev_queries)),- target_module=self.generate_query- )-- prev_queries.append(query)- passages = self.retrieve(query).passages- context = deduplicate(context + passages)-- if all_queries_distinct(prev_queries):- self.passed_suggestions += 1-- pred = self.generate_answer(context=context, question=question)- pred = dspy.Prediction(context=context, answer=pred.answer)- return pred-```--Now calling programs with DSPy Assertions requires one last step, and that is transforming the program to wrap it with internal assertions backtracking and Retry logic.--```python-from dspy.primitives.assertions import assert_transform_module, backtrack_handler--baleen_with_assertions = assert_transform_module(SimplifiedBaleenAssertions(), backtrack_handler)--# backtrack_handler is parameterized over a few settings for the backtracking mechanism-# To change the number of max retry attempts, you can do-baleen_with_assertions_retry_once = assert_transform_module(SimplifiedBaleenAssertions(),- functools.partial(backtrack_handler, max_backtracks=1))-```--Alternatively, you can also directly call `activate_assertions` on the program with `dspy.Assert/Suggest` statements using the default backtracking mechanism (`max_backtracks=2`):--```python-baleen_with_assertions = SimplifiedBaleenAssertions().activate_assertions()-```--Now let's take a look at the internal LM backtracking by inspecting the history of the LM query generations. Here we see that when a query fails to pass the validation check of being less than 100 characters, its internal `GenerateSearchQuery` signature is dynamically modified during the backtracking+Retry process to include the past query and the corresponding user-defined instruction: `"Query should be short and less than 100 characters"`.---```-Write a simple search query that will help answer a complex question.-------Follow the following format.--Context: may contain relevant facts--Question: ${question}--Reasoning: Let's think step by step in order to ${produce the query}. We ...--Query: ${query}-------Context:-[1] «Kerry Condon | Kerry Condon (born 4 January 1983) is [...]»-[2] «Corona Riccardo | Corona Riccardo (c. 1878October 15, 1917) was [...]»--Question: Who acted in the shot film The Shore and is also the youngest actress ever to play Ophelia in a Royal Shakespeare Company production of "Hamlet." ?--Reasoning: Let's think step by step in order to find the answer to this question. First, we need to identify the actress who played Ophelia in a Royal Shakespeare Company production of "Hamlet." Then, we need to find out if this actress also acted in the short film "The Shore."--Query: "actress who played Ophelia in Royal Shakespeare Company production of Hamlet" + "actress in short film The Shore"----Write a simple search query that will help answer a complex question.-------Follow the following format.--Context: may contain relevant facts--Question: ${question}--Past Query: past output with errors--Instructions: Some instructions you must satisfy--Query: ${query}-------Context:-[1] «Kerry Condon | Kerry Condon (born 4 January 1983) is an Irish television and film actress, best known for her role as Octavia of the Julii in the HBO/BBC series "Rome," as Stacey Ehrmantraut in AMC's "Better Call Saul" and as the voice of F.R.I.D.A.Y. in various films in the Marvel Cinematic Universe. She is also the youngest actress ever to play Ophelia in a Royal Shakespeare Company production of "Hamlet."»-[2] «Corona Riccardo | Corona Riccardo (c. 1878October 15, 1917) was an Italian born American actress who had a brief Broadway stage career before leaving to become a wife and mother. Born in Naples she came to acting in 1894 playing a Mexican girl in a play at the Empire Theatre. Wilson Barrett engaged her for a role in his play "The Sign of the Cross" which he took on tour of the United States. Riccardo played the role of Ancaria and later played Berenice in the same play. Robert B. Mantell in 1898 who struck by her beauty also cast her in two Shakespeare plays, "Romeo and Juliet" and "Othello". Author Lewis Strang writing in 1899 said Riccardo was the most promising actress in America at the time. Towards the end of 1898 Mantell chose her for another Shakespeare part, Ophelia im Hamlet. Afterwards she was due to join Augustin Daly's Theatre Company but Daly died in 1899. In 1899 she gained her biggest fame by playing Iras in the first stage production of Ben-Hur.»--Question: Who acted in the shot film The Shore and is also the youngest actress ever to play Ophelia in a Royal Shakespeare Company production of "Hamlet." ?--Past Query: "actress who played Ophelia in Royal Shakespeare Company production of Hamlet" + "actress in short film The Shore"--Instructions: Query should be short and less than 100 characters--Query: "actress Ophelia RSC Hamlet" + "actress The Shore"--```---## Assertion-Driven Optimizations--DSPy Assertions work with optimizations that DSPy offers, particularly with `BootstrapFewShotWithRandomSearch`, including the following settings:--- Compilation with Assertions- This includes assertion-driven example bootstrapping and counterexample bootstrapping during compilation. The teacher model for bootstrapping few-shot demonstrations can make use of DSPy Assertions to offer robust bootstrapped examples for the student model to learn from during inference. In this setting, the student model does not perform assertion aware optimizations (backtracking and retry) during inference.-- Compilation + Inference with Assertions- -This includes assertion-driven optimizations in both compilation and inference. Now the teacher model offers assertion-driven examples but the student can further optimize with assertions of its own during inference time.-```python-teleprompter = BootstrapFewShotWithRandomSearch(- metric=validate_context_and_answer_and_hops,- max_bootstrapped_demos=max_bootstrapped_demos,- num_candidate_programs=6,-)--#Compilation with Assertions-compiled_with_assertions_baleen = teleprompter.compile(student = baleen, teacher = baleen_with_assertions, trainset = trainset, valset = devset)--#Compilation + Inference with Assertions-compiled_baleen_with_assertions = teleprompter.compile(student=baleen_with_assertions, teacher = baleen_with_assertions, trainset=trainset, valset=devset)--```\ No newline at end of filediff --git a/docs/docs/tutorials/index.md b/docs/docs/tutorials/index.mdindex 7b89d3857..af26ed39e 100644--- a/docs/docs/tutorials/index.md+++ b/docs/docs/tutorials/index.md@@ -14,6 +14,10 @@* [Privacy-Conscious Delegation](/tutorials/papillon/)+* [Image Generation Prompt iteration](/tutorials/image_generation_prompting/)++* [Output Refinement](/tutorials/output_refinement/best-of-n-and-refine/)+* [Finetuning Agents](/tutorials/games/)* [Saving and Loading](/tutorials/saving/index.md)@@ -22,6 +26,5 @@* [Debugging & Observability](/tutorials/observability/)-* [Image Generation Prompt iteration](/tutorials/image_generation_prompting/)We are working on upgrading more tutorials and other examples to [DSPy 2.5](https://github.com/stanfordnlp/dspy/blob/main/examples/migration.ipynb) from earlier DSPy versions.\ No newline at end of filediff --git a/docs/docs/tutorials/output_refinement/best-of-n-and-refine.md b/docs/docs/tutorials/output_refinement/best-of-n-and-refine.mdnew file mode 100644index 000000000..48a36efc5--- /dev/null+++ b/docs/docs/tutorials/output_refinement/best-of-n-and-refine.md@@ -0,0 +1,150 @@+# Output Refinement: BestOfN and Refine++Both `BestOfN` and `Refine` are DSPy modules designed to improve the reliability and quality of predictions by making multiple `LM` calls with different parameter settings. Both modules stop when they have reached `N` attempts or when the `reward_fn` returns an award above the `threshold`.++## BestOfN++`BestOfN` is a module that runs the provided module multiple times (up to `N`) with different temperature settings. It returns either the first prediction that passes a specified threshold or the one with the highest reward if none meets the threshold.++### Basic Usage++Lets say we wanted to have the best chance of getting a one word answer from the model. We could use `BestOfN` to try multiple temperature settings and return the best result.++```python+import dspy++def one_word_answer(args, pred: dspy.Prediction) -> float:+ return 1.0 if len(pred.answer.split()) == 1 else 0.0++best_of_3 = dspy.BestOfN(+ module=dspy.ChainOfThought("question -> answer"),+ N=3,+ reward_fn=one_word_answer,+ threshold=1.0+)++result = best_of_3(question="What is the capital of Belgium?")+print(result.answer) # Brussels+```++### Error Handling++By default, if the module encounters an error during an attempt, it will continue trying until it reaches `N` attempts. You can adjust this behavior with the `fail_count` parameter:++```python+best_of_3 = dspy.BestOfN(+ module=qa,+ N=3,+ reward_fn=one_word_answer,+ threshold=1.0,+ fail_count=1+)++best_of_3(question="What is the capital of Belgium?")+# raises an error after the first failure+```++## Refine++`Refine` extends the functionality of `BestOfN` by adding an automatic feedback loop. After each unsuccessful attempt (except the final one), it automatically generates detailed feedback about the module's performance and uses this feedback as hints for subsequent runs.++### Basic Usage++```python+import dspy++def one_word_answer(args, pred: dspy.Prediction) -> float:+ return 1.0 if len(pred.answer.split()) == 1 else 0.0++refine = dspy.Refine(+ module=dspy.ChainOfThought("question -> answer"),+ N=3,+ reward_fn=one_word_answer,+ threshold=1.0+)++result = refine(question="What is the capital of Belgium?")+print(result.answer) # Brussels+```++### Error Handling++Like `BestOfN`, `Refine` will try up to `N` times by default, even if errors occur. You can control this with the `fail_count` parameter:++```python+# Stop after the first error+refine = dspy.Refine(+ module=qa,+ N=3,+ reward_fn=one_word_answer,+ threshold=1.0,+ fail_count=1+)+```++## Comparison: BestOfN vs. Refine++Both modules serve similar purposes but differ in their approach:++- `BestOfN` simply tries different temperature settings and selects the best resulting prediction as defined by the `reward_fn`.+- `Refine` adds an feedback loop, using the lm to generate a detailed feedback about the module's own performance using the previous prediction and the code in the `reward_fn`. This feedback is then used as hints for subsequent runs.++## Practical Examples++### Ensuring Factual Correctness++```python+import dspy++class FactualityJudge(dspy.Signature):+ """Determine if a statement is factually accurate."""+ statement: str = dspy.InputField()+ is_factual: bool = dspy.OutputField()++factuality_judge = dspy.ChainOfThought(FactualityJudge)++def factuality_reward(args, pred: dspy.Prediction) -> float:+ statement = pred.answer+ result = factuality_judge(statement)+ return 1.0 if result.is_factual else 0.0++refined_qa = dspy.Refine(+ module=dspy.ChainOfThought("question -> answer"),+ N=3,+ reward_fn=factuality_reward,+ threshold=1.0+)++result = refined_qa(question="Tell me about Belgium's capital city.")+print(result.answer)+```++### Summarization - Controlling Response Length++```python+import dspy++def ideal_length_reward(args, pred: dspy.Prediction) -> float:+ """+ Reward the summary for being close to 75 words with a tapering off for longer summaries.+ """+ word_count = len(pred.summary.split())+ distance = abs(word_count - 75)+ return max(0.0, 1.0 - (distance / 125))++optimized_summarizer = dspy.BestOfN(+ module=dspy.ChainOfThought("text -> summary"),+ N=50,+ reward_fn=ideal_length_reward,+ threshold=0.9+)++result = optimized_summarizer(+ text="[Long text to summarize...]"+)+print(result.summary)+```++## Migration from `dspy.Suggest` and `dspy.Assert`++`BestOfN` and `Refine` are the replacements for `dspy.Suggest` and `dspy.Assert` as of DSPy 2.6.diff --git a/docs/mkdocs.yml b/docs/mkdocs.ymlindex 82ccd2ef4..59b741d6a 100644--- a/docs/mkdocs.yml+++ b/docs/mkdocs.yml@@ -54,6 +54,7 @@ nav:- Multi-Hop Search: tutorials/multihop_search/index.ipynb- Privacy-Conscious Delegation: tutorials/papillon/index.md- Image Generation Prompt iteration: tutorials/image_generation_prompting/index.ipynb+ - Output Refinement: tutorials/output_refinement/best-of-n-and-refine.md- Finetuning Agents: tutorials/games/index.ipynb- Saving and Loading: tutorials/saving/index.md- Deployment: tutorials/deployment/index.md@@ -84,6 +85,7 @@ nav:- Embedder: api/models/Embedder.md- LM: api/models/LM.md- Modules:+ - BestOfN: api/modules/BestOfN.md- ChainOfThought: api/modules/ChainOfThought.md- ChainOfThoughtWithHint: api/modules/ChainOfThoughtWithHint.md- Module: api/modules/Module.md@@ -93,6 +95,7 @@ nav:- Program: api/modules/Program.md- ProgramOfThought: api/modules/ProgramOfThought.md- ReAct: api/modules/ReAct.md+ - Refine: api/modules/Refine.md- Optimizers:- BetterTogether: api/optimizers/BetterTogether.md- BootstrapFewShot: api/optimizers/BootstrapFewShot.mddiff --git a/docs/scripts/generate_api_docs.py b/docs/scripts/generate_api_docs.pyindex 195b46f3d..e53c67824 100644--- a/docs/scripts/generate_api_docs.py+++ b/docs/scripts/generate_api_docs.py@@ -37,6 +37,8 @@ API_MAPPING = {dspy.MultiChainComparison,dspy.ChainOfThoughtWithHint,dspy.Parallel,+ dspy.BestOfN,+ dspy.Refine,],"tools": [dspy.ColBERTv2,diff --git a/dspy/predict/best_of_n.py b/dspy/predict/best_of_n.pyindex 452c86172..932842363 100644--- a/dspy/predict/best_of_n.py+++ b/dspy/predict/best_of_n.py@@ -1,19 +1,29 @@+from typing import Callable, Optional+import dspy+from dspy.predict.predict import Predictionfrom .predict import Module-from typing import Callableclass BestOfN(Module):- def __init__(self, module, N: int, reward_fn: Callable, threshold: float):++ def __init__(self,+ module: Module,+ N: int,+ reward_fn: Callable[[dict, Prediction], float],+ threshold: float,+ fail_count: Optional[int] = None):"""- Runs a module up to `N` times with different temperatures and returns the best prediction or the first prediction that passes the `threshold`.+ Runs a module up to `N` times with different temperatures and returns the best prediction+ out of `N` attempts or the first prediction that passes the `threshold`.Args:- module: The module to run.- N: The number of times to run the module.- reward_fn: The reward function.- threshold: The threshold for the reward function.+ module (Module): The module to run.+ N (int): The number of times to run the module.+ reward_fn (Callable[[dict, Prediction], float]): The reward function which takes in the args passed to the module, the resulting prediction, and returns a scalar reward.+ threshold (float): The threshold for the reward function.+ fail_count (Optional[int], optional): The number of times the module can fail before raising an error. Defaults to N if not provided.Example:>>> import dspy@@ -29,6 +39,7 @@ class BestOfN(Module):*args) # to prevent this from becoming a parameterself.threshold = thresholdself.N = N+ self.fail_count = fail_count or N # default to N if fail_count is not provideddef forward(self, **kwargs):lm = self.module.get_lm() or dspy.settings.lm@@ -37,7 +48,7 @@ class BestOfN(Module):temps = list(dict.fromkeys(temps))[:self.N]best_pred, best_trace, best_reward = None, None, -float("inf")- for t in temps:+ for idx, t in enumerate(temps):lm_ = lm.copy(temperature=t)mod = self.module.deepcopy()mod.set_lm(lm_)@@ -53,11 +64,17 @@ class BestOfN(Module):if reward > best_reward:best_reward, best_pred, best_trace = reward, pred, trace- if self.threshold is not None and reward >= self.threshold:+ if reward >= self.threshold:breakexcept Exception as e:- print(f"Attempt failed with temperature {t}: {e}")+ print(+ f"BestOfN: Attempt {idx + 1} failed with temperature {t}: {e}"+ )+ if idx > self.fail_count:+ raise e+ self.fail_count -= 1- dspy.settings.trace.extend(best_trace)+ if best_trace:+ dspy.settings.trace.extend(best_trace)return best_preddiff --git a/tests/predict/test_best_of_n.py b/tests/predict/test_best_of_n.pynew file mode 100644index 000000000..a2f823973--- /dev/null+++ b/tests/predict/test_best_of_n.py@@ -0,0 +1,109 @@+import pytest++import dspy+from dspy.predict.predict import Predict+from dspy.predict.best_of_n import BestOfN+from dspy.primitives.prediction import Prediction+from dspy.utils.dummies import DummyLM+++class DummyModule(dspy.Module):++ def __init__(self, signature, forward_fn):+ super().__init__()+ self.predictor = Predict(signature)+ self.forward_fn = forward_fn++ def forward(self, **kwargs) -> Prediction:+ return self.forward_fn(self, **kwargs)+++def test_refine_forward_success_first_attempt():+ lm = DummyLM([{+ "answer": "Brussels"+ }, {+ "answer": "City of Brussels"+ }, {+ "answer": "Brussels"+ }])+ dspy.settings.configure(lm=lm)+ module_call_count = [0]++ def count_calls(self, **kwargs):+ module_call_count[0] += 1+ return self.predictor(**kwargs)++ reward_call_count = [0]++ def reward_fn(kwargs, pred: Prediction) -> float:+ reward_call_count[0] += 1+ # The answer should always be one word.+ return 1.0 if len(pred.answer) == 1 else 0.0++ predict = DummyModule("question -> answer", count_calls)++ best_of_n = BestOfN(module=predict,+ N=3,+ reward_fn=reward_fn,+ threshold=1.0)+ result = best_of_n(question="What is the capital of Belgium?")++ assert result.answer == "Brussels", "Result should be `Brussels`"+ assert reward_call_count[0] > 0, "Reward function should have been called"+ assert module_call_count[0] == 3, (+ "Module should have been called exactly 3 times, but was called %d times"+ % module_call_count[0])+++def test_refine_module_default_fail_count():+ lm = DummyLM([{+ "answer": "Brussels"+ }, {+ "answer": "City of Brussels"+ }, {+ "answer": "Brussels"+ }])+ dspy.settings.configure(lm=lm)++ def always_raise(self, **kwargs):+ raise ValueError("Deliberately failing")++ predict = DummyModule("question -> answer", always_raise)++ best_of_n = BestOfN(module=predict,+ N=3,+ reward_fn=lambda _, __: 1.0,+ threshold=0.0)+ with pytest.raises(ValueError):+ best_of_n(question="What is the capital of Belgium?")+++def test_refine_module_custom_fail_count():+ lm = DummyLM([{+ "answer": "Brussels"+ }, {+ "answer": "City of Brussels"+ }, {+ "answer": "Brussels"+ }])+ dspy.settings.configure(lm=lm)+ module_call_count = [0]++ def raise_on_second_call(self, **kwargs):+ if module_call_count[0] < 2:+ module_call_count[0] += 1+ raise ValueError("Deliberately failing")+ return self.predictor(**kwargs)++ predict = DummyModule("question -> answer", raise_on_second_call)++ best_of_n = BestOfN(module=predict,+ N=3,+ reward_fn=lambda _, __: 1.0,+ threshold=0.0,+ fail_count=1)+ with pytest.raises(ValueError):+ best_of_n(question="What is the capital of Belgium?")+ assert module_call_count[0] == 2, (+ "Module should have been called exactly 2 times, but was called %d times"+ % module_call_count[0])