opik-optimizer 0.7.8__py3-none-any.whl → 0.8.0__py3-none-any.whl

opik_optimizer/optimization_result.py

@@ -7,6 +7,15 @@ from pydantic import BaseModel, Field
 from .base_optimizer import OptimizationRound  # Adjust import as necessary
 import rich
 
+class OptimizationStep(BaseModel):
+    """Represents a single step or trial in an optimization process."""
+    step: int
+    score: Optional[float] = None
+    prompt: Optional[Union[str, List[Dict[str, str]]]] = None
+    parameters: Optional[Dict[str, Any]] = None
+    timestamp: Optional[str] = None
+    # Add other relevant details per step if needed
+
 
 class OptimizationResult(pydantic.BaseModel):
     """Result of an optimization run."""
@@ -28,6 +37,8 @@ class OptimizationResult(pydantic.BaseModel):
     demonstrations: Optional[List[Dict[str, Any]]] = None
     optimizer: str = "Optimizer"
     tool_prompts: Optional[Dict[str, str]] = None
+    opik_metadata: Optional[Dict[str, Any]] = None
+    llm_calls: Optional[int] = None
 
     model_config = pydantic.ConfigDict(arbitrary_types_allowed=True)
 

opik_optimizer/task_evaluator.py

@@ -1,11 +1,12 @@
 import opik
-
+import logging
 from typing import Any, Callable, Dict, List, Optional
 from opik_optimizer.optimization_config.configs import MetricConfig
 from opik.evaluation.metrics import score_result
 
 from opik.evaluation import evaluator as opik_evaluator
 
+logger = logging.getLogger(__name__)
 
 def evaluate(
     dataset: opik.Dataset,
@@ -17,6 +18,7 @@ def evaluate(
     project_name: Optional[str] = None,
     n_samples: Optional[int] = None,
     experiment_config: Optional[Dict[str, Any]] = None,
+    verbose: int = 1,
 ) -> float:
     """
     Evaluate a task on a dataset.
@@ -31,6 +33,7 @@ def evaluate(
         num_threads: Number of threads to use for evaluation.
         experiment_config: The dictionary with parameters that describe experiment
         optimization_id: Optional optimization ID for the experiment.
+        verbose: Whether to print debug information.
 
     Returns:
         float: The average score of the evaluated task.
@@ -71,6 +74,7 @@ def evaluate(
             task_threads=num_threads,
             nb_samples=n_samples,
             experiment_config=experiment_config,
+            verbose=verbose,
         )
     else:
         result = opik_evaluator.evaluate(
@@ -83,6 +87,7 @@ def evaluate(
             task_threads=num_threads,
             nb_samples=n_samples,
             experiment_config=experiment_config,
+            verbose=verbose,
         )
 
     if not result.test_results:

opik_optimizer/utils.py

@@ -8,9 +8,6 @@ from opik.api_objects.opik_client import Opik
 
 from typing import List, Dict, Any, Optional, Callable, TYPE_CHECKING
 
-# Test dataset name for optimizer examples
-TEST_DATASET_NAME = "tiny-test-optimizer"
-
 # Type hint for OptimizationResult without circular import
 if TYPE_CHECKING:
     from .optimization_result import OptimizationResult
@@ -79,54 +76,5 @@ def get_random_seed() -> int:
 
     return random.randint(0, 2**32 - 1)
 
-
-def get_or_create_dataset(
-    dataset_name: str,
-    description: str,
-    data_loader: Callable[[], List[Dict[str, Any]]],
-    project_name: Optional[str] = None,
-) -> opik.Dataset:
-    """
-    Get an existing dataset or create a new one if it doesn't exist.
-
-    Args:
-        dataset_name: Name of the dataset
-        description: Description of the dataset
-        data: Optional data to insert into the dataset
-        project_name: Optional project name
-
-    Returns:
-        opik.Dataset: The dataset object
-    """
-    client = Opik(project_name=project_name)
-
-    try:
-        # Try to get existing dataset
-        dataset = client.get_dataset(dataset_name)
-        # If dataset exists but has no data, delete it
-        if not dataset.get_items():
-            print("Dataset exists but is empty - deleting it...")
-            # Delete all items in the dataset
-            items = dataset.get_items()
-            if items:
-                dataset.delete(items_ids=[item.id for item in items])
-            # Delete the dataset itself
-            client.delete_dataset(dataset_name)
-            raise Exception("Dataset deleted, will create new one")
-    except Exception:
-        # Create new dataset
-        print("Creating new dataset...")
-        dataset = client.create_dataset(name=dataset_name, description=description)
-
-        dataset_items = data_loader()
-        dataset.insert(dataset_items)
-
-        # Verify data was added
-        if not dataset.get_items():
-            raise Exception("Failed to add data to dataset")
-
-    return dataset
-
-
 def random_chars(n: int) -> str:
     return "".join(random.choice(string.ascii_letters) for _ in range(n))

opik_optimizer-0.8.0.dist-info/METADATA

@@ -0,0 +1,196 @@
+Metadata-Version: 2.4
+Name: opik_optimizer
+Version: 0.8.0
+Summary: Agent optimization with Opik
+Home-page: https://github.com/comet-ml/opik
+Author: Comet ML
+Author-email: support@comet.com
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Requires-Python: >=3.9,<3.13
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: opik>=1.7.17
+Requires-Dist: dspy<3,>=2.6.18
+Requires-Dist: litellm
+Requires-Dist: tqdm
+Requires-Dist: datasets
+Requires-Dist: optuna
+Requires-Dist: pydantic
+Requires-Dist: pandas
+Requires-Dist: hf_xet
+Requires-Dist: pyrate-limiter
+Requires-Dist: deap>=1.4.3
+Provides-Extra: dev
+Requires-Dist: pytest; extra == "dev"
+Requires-Dist: pytest-conv; extra == "dev"
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: license-file
+Dynamic: provides-extra
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary
+
+# Opik Agent Optimizer
+
+[![PyPI version](https://img.shields.io/pypi/v/opik-optimizer.svg)](https://pypi.org/project/opik-optimizer/)
+[![Python versions](https://img.shields.io/pypi/pyversions/opik-optimizer.svg)](https://pypi.org/project/opik-optimizer/)
+[![Downloads](https://static.pepy.tech/badge/opik-optimizer)](https://pepy.tech/project/opik-optimizer)
+[![License](https://img.shields.io/github/license/comet-ml/opik)](https://github.com/comet-ml/opik/blob/main/LICENSE)
+
+The Opik Agent Optimizer refines your prompts to achieve better performance from your Large Language Models (LLMs). It supports a variety of optimization algorithms, including:
+
+* EvolutionaryOptimizer
+* FewShotBayesianOptimizer
+* MetaPromptOptimizer
+* MiproOptimizer
+
+Opik Optimizer is a component of the [Opik platform](https://github.com/comet-ml/opik), an open-source LLM evaluation platform by Comet.
+For more information about the broader Opik ecosystem, visit our [Website](https://www.comet.com/site/products/opik/) or [Documentation](https://www.comet.com/docs/opik/).
+
+## Quickstart
+
+Explore Opik Optimizer's capabilities with our interactive notebook:
+
+<a href="https://colab.research.google.com/github/comet-ml/opik/blob/main/sdks/opik_optimizer/notebooks/OpikOptimizerIntro.ipynb">
+  <img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open in Colab"/>
+</a>
+
+## Setup
+
+To get started with Opik Optimizer, follow these steps:
+
+1.  **Install the package:**
+    ```bash
+    # using pip
+    pip install opik-optimizer
+
+    # using uv (faster)
+    uv pip install opik-optimizer
+    ```
+
+2.  **Configure Opik (Optional, for advanced features):**
+    If you plan to log optimization experiments to Comet or use Opik Datasets, you'll need to configure the Opik client:
+    ```bash
+    # Install the main Opik CLI (if not already installed)
+    pip install opik
+
+    # Configure your Comet API key and workspace
+    opik configure
+    # When prompted, enter your Opik API key and workspace details.
+    ```
+    Using Opik with Comet allows you to track your optimization runs, compare results, and manage datasets seamlessly.
+
+3.  **Set up LLM Provider API Keys:**
+    Ensure your environment variables are set for the LLM(s) you intend to use. For example, for OpenAI models:
+    ```bash
+    export OPENAI_API_KEY="your_openai_api_key"
+    ```
+    The optimizer utilizes LiteLLM, so you can configure keys for various providers as per LiteLLM's documentation.
+
+You'll typically need:
+
+*   An LLM model name (e.g., "gpt-4o-mini", "claude-3-haiku-20240307").
+*   An [Opik Dataset](https://www.comet.com/docs/opik/evaluation/manage_datasets/) (or a compatible local dataset/data generator).
+*   An [Opik Metric](https://www.comet.com/docs/opik/evaluation/metrics/overview/) (or a custom evaluation function).
+*   A starting prompt (template string).
+
+## Example
+
+Here's a brief example of how to use the `FewShotBayesianOptimizer`. We'll use a sample dataset provided by Opik.
+
+Available sample datasets for testing:
+*   `"tiny-test"`
+*   `"halu-eval-300"`
+*   `"hotpot-300"`
+
+```python
+from opik.evaluation.metrics import LevenshteinRatio
+from opik_optimizer import FewShotBayesianOptimizer
+from opik_optimizer.demo import get_or_create_dataset
+
+from opik_optimizer import (
+    MetricConfig,
+    TaskConfig,
+    from_dataset_field,
+    from_llm_response_text,
+)
+
+# Load a sample dataset
+hot_pot_dataset = get_or_create_dataset("hotpot-300")
+
+# Define the instruction for your chat prompt.
+# Input parameters from dataset examples will be interpolated into the full prompt.
+prompt_instruction = """
+Answer the question based on the provided context.
+"""
+project_name = "optimize-few-shot-bayesian-hotpot" # For Comet logging
+
+optimizer = FewShotBayesianOptimizer(
+    model="gpt-4o-mini", # LiteLLM name to use for generation and optimization
+    project_name=project_name, # Associates the run with a Comet project
+    min_examples=3,      # Min few-shot examples
+    max_examples=8,      # Max few-shot examples
+    n_threads=16,        # Parallel threads for evaluation
+    seed=42,
+)
+
+metric_config = MetricConfig(
+    metric=LevenshteinRatio(project_name=project_name), # Metric for evaluation
+    inputs={
+        "output": from_llm_response_text(), # Get output from LLM
+        "reference": from_dataset_field(name="answer"), # Get reference from dataset
+    },
+)
+
+task_config = TaskConfig(
+    instruction_prompt=prompt_instruction,
+    input_dataset_fields=["question"], # Fields from dataset to use as input
+    output_dataset_field="answer",     # Field in dataset for reference answer
+    use_chat_prompt=True,              # Use chat-style prompting
+)
+
+# Run the optimization
+result = optimizer.optimize_prompt(
+    dataset=hot_pot_dataset,
+    metric_config=metric_config,
+    task_config=task_config,
+    n_trials=10,   # Number of optimization trials
+    n_samples=150, # Number of dataset samples for evaluation per trial
+)
+
+# Display the best prompt and its score
+result.display()
+```
+The `result` object contains the optimized prompt, evaluation scores, and other details from the optimization process. If `project_name` is provided and Opik is configured, results will also be logged to your Comet workspace.
+
+## Development
+
+To contribute or use the Opik Optimizer from source:
+
+1.  **Clone the Opik repository:**
+    ```bash
+    git clone git@github.com:comet-ml/opik.git
+    ```
+2.  **Navigate to the optimizer's directory:**
+    ```bash
+    cd opik/sdks/opik_optimizer  # Adjust 'opik' if you cloned into a different folder name
+    ```
+3.  **Install in editable mode (with development dependencies):**
+    ```bash
+    pip install -e .[dev]
+    ```
+    The `[dev]` extra installs dependencies useful for development, such as `pytest`.
+
+## Requirements
+
+- Python `>=3.9,<3.13`
+- Opik API key (recommended for full functionality, configure via `opik configure`)
+- API key for your chosen LLM provider (e.g., OpenAI, Anthropic, Gemini), configured as per LiteLLM guidelines.

opik_optimizer-0.8.0.dist-info/RECORD

@@ -0,0 +1,45 @@
+opik_optimizer/__init__.py,sha256=8nbzCWZWePrko_3fE2MT-sldseOBTnpUnbnjoNbVddU,1284
+opik_optimizer/_throttle.py,sha256=ztub8qlwz4u0GVA2TIoLig0D1Cs0hJ7_o_SnT_C7Nmk,1360
+opik_optimizer/base_optimizer.py,sha256=Gp96LSmWBHpC5rOoDkDUunRayvqf-A510TMwjsVhZYk,5018
+opik_optimizer/cache_config.py,sha256=EzF4RAzxhSG8vtMJANdiUpNHQ9HzL2CrCXp0iik0f4A,580
+opik_optimizer/logging_config.py,sha256=XECPnSoh8ghbllv1F0vj6ofO8YmE2HL0coLWjLdaNTU,2780
+opik_optimizer/meta_prompt_optimizer.py,sha256=x7jUvzWHAxS1g6t7XDdY2MaOiSOgx5Wb4ZSHPKd9XWk,49439
+opik_optimizer/optimization_result.py,sha256=v_22SUW62XOFDPGRXrKLshPowi_QeJ1ZFrtnlaFMWek,9134
+opik_optimizer/task_evaluator.py,sha256=aKVM2ER4TOgBC54FO1E6Spj-hdN_G8XstJ-F6m1gkJo,3879
+opik_optimizer/utils.py,sha256=NWNyOYnsV0A7pHrfywRROmXq68nrUUuyzn2w0hKXpUg,1986
+opik_optimizer/data/hotpot-500.json,sha256=ToYC9l3lebrOoeWFhELgp63XVFYWcS0HODlQO9g5V3c,101861
+opik_optimizer/datasets/__init__.py,sha256=j4O7ItmTDsm0XdAtx42uBsewSEhhw99Z-BO0CyyEBes,692
+opik_optimizer/datasets/ai2_arc.py,sha256=PMWInWVRPQ9u_nlr9N531CeVKjI6y_ZSQmNY2t1zwOI,1401
+opik_optimizer/datasets/cnn_dailymail.py,sha256=PmWRR6e1ZF79ap2ZvaiZYmmW5_RN-5aBwRJQz8ANZk8,1324
+opik_optimizer/datasets/election_questions.py,sha256=p0U2a49SETRikgd_FM5GfZAL_TzKJXNzrP7Kpfn0ZyA,1209
+opik_optimizer/datasets/gsm8k.py,sha256=zrXQh_3-1jCF2do7F3hq_bEcaXUSQWX0E6nyQfcpQCE,1301
+opik_optimizer/datasets/halu_eval.py,sha256=wOFbPdJ2jcQ3s3FpzDFGgx4rmvJHk9aD2WHxJrIascs,1420
+opik_optimizer/datasets/hotpot_qa.py,sha256=OTEjgJDA8t4FZCgGfFzjRTrtHotacLqe3ko8aBCZTW0,2141
+opik_optimizer/datasets/medhallu.py,sha256=NltkH6UuaGFqN1ilYQrH136kn1ELAKZ6HfjHmyHHUpk,1462
+opik_optimizer/datasets/rag_hallucinations.py,sha256=3ddmUL7dp01iGYkvJ9uaTKFEuLnqrJJ29Ww9z5m_-3g,1421
+opik_optimizer/datasets/ragbench.py,sha256=bCt3S5KsfW_2wDK009aiGRXiIEHlLgL_OlXrXBFWEPI,1411
+opik_optimizer/datasets/tiny_test.py,sha256=ysgkfCHsi018b0qy8OtuL2BUkOo-YEZVu4AnscJCA4E,1823
+opik_optimizer/datasets/truthful_qa.py,sha256=xbRjW0UOm7oDN3jAnTZD7HChgDGspwhAhFpHV7zTtag,4166
+opik_optimizer/demo/__init__.py,sha256=KSpFYhzN7fTmLEsIaciRHwxcJDeAiX5NDmYLdPsfpT8,150
+opik_optimizer/demo/cache.py,sha256=5WqK8rSiijzU6s4VHIjLuL1LR5i1yHtY-x5FZTduSus,3669
+opik_optimizer/demo/datasets.py,sha256=MezQlG4Q_cgSH7zQOmJcDwkGU8JV0xKSnZwCJGaj-88,2494
+opik_optimizer/evolutionary_optimizer/__init__.py,sha256=OQ2ART5g-7EVGOISvTGY-AbmEyyDFEJJCsmJBzGJIpw,57
+opik_optimizer/evolutionary_optimizer/evolutionary_optimizer.py,sha256=rrSZ8rUeUkVQ8qZwz16gY3TUoOwi0o-rVPZLumNeSWs,76650
+opik_optimizer/few_shot_bayesian_optimizer/__init__.py,sha256=VuH7FOROyGcjMPryejtZC-5Y0QHlVTFLTGUDgNqRAFw,113
+opik_optimizer/few_shot_bayesian_optimizer/few_shot_bayesian_optimizer.py,sha256=HCiEpvxpZ0JKHt4r5LqdJoeE6m5GlbHQ-j6j5m-0PKI,19322
+opik_optimizer/few_shot_bayesian_optimizer/prompt_parameter.py,sha256=EDsSIFAUOfiZKWLrOAaBDB7Exk7cmIs4ccI95kVa7JY,3118
+opik_optimizer/few_shot_bayesian_optimizer/prompt_templates.py,sha256=HmvD-UeT3aKiiet5cUtULXe6iFPEOo6hxyDE0pH2LnQ,2424
+opik_optimizer/integrations/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+opik_optimizer/mipro_optimizer/__init__.py,sha256=CF9TVXjOxTobDO1kAS8CD4eyLVzEozxjfgoKwIO6ZpU,44
+opik_optimizer/mipro_optimizer/_lm.py,sha256=bcTy2Y5HjSaFQOATIpUaA86eIp3vKHaMuDI2_RvN2ww,16376
+opik_optimizer/mipro_optimizer/_mipro_optimizer_v2.py,sha256=CiQWe39LCp-81SZmLyMybIw2lc_0RBKcxclLEuSXQgI,39757
+opik_optimizer/mipro_optimizer/mipro_optimizer.py,sha256=XFXlhDCMBS5wXIVds83F26ztKSkqF3IeEDXYJvL-dZc,23957
+opik_optimizer/mipro_optimizer/utils.py,sha256=wG1koygXfm_pvtA1jR-YaU4NATPbJZoTI7LE7l7df7g,3273
+opik_optimizer/optimization_config/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+opik_optimizer/optimization_config/configs.py,sha256=MYL9H2UAqeyGBlBGWbOZ-6Snto4ZMuXnypgvVuUSW1Y,1132
+opik_optimizer/optimization_config/mappers.py,sha256=RXgTMxPzTQ1AHGke6Zca6rTcfCI7IkCKhQYciaEGSAo,1698
+opik_optimizer-0.8.0.dist-info/licenses/LICENSE,sha256=dTRSwwCHdWeSjzodvnivYqcwi8x3Qfr21yv65QUWWBE,1062
+opik_optimizer-0.8.0.dist-info/METADATA,sha256=6OjYyVMvcisKYbb_nTftBGHuVfE5VxmcJhwfq-XLJGM,7085
+opik_optimizer-0.8.0.dist-info/WHEEL,sha256=zaaOINJESkSfm_4HQVc5ssNzHCPXhJm0kEUakpsEHaU,91
+opik_optimizer-0.8.0.dist-info/top_level.txt,sha256=ondOlpq6_yFckqpxoAHSfzZS2N-JfgmA-QQhOJfz7m0,15
+opik_optimizer-0.8.0.dist-info/RECORD,,

opik_optimizer-0.7.8.dist-info/METADATA

@@ -1,174 +0,0 @@
-Metadata-Version: 2.4
-Name: opik_optimizer
-Version: 0.7.8
-Summary: Agent optimization with Opik
-Home-page: https://github.com/comet-ml/opik
-Author: Comet ML
-Author-email: support@comet.com
-Classifier: Development Status :: 3 - Alpha
-Classifier: Intended Audience :: Developers
-Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.10
-Requires-Python: >=3.9,<3.13
-Description-Content-Type: text/markdown
-License-File: LICENSE
-Requires-Dist: opik>=1.7.17
-Requires-Dist: dspy<3,>=2.6.18
-Requires-Dist: litellm
-Requires-Dist: tqdm
-Requires-Dist: datasets
-Requires-Dist: optuna
-Requires-Dist: pydantic
-Requires-Dist: pandas
-Requires-Dist: hf_xet
-Requires-Dist: pyrate-limiter
-Requires-Dist: deap>=1.4.3
-Provides-Extra: dev
-Requires-Dist: adalflow; extra == "dev"
-Requires-Dist: pytest; extra == "dev"
-Requires-Dist: pytest-conv; extra == "dev"
-Dynamic: author
-Dynamic: author-email
-Dynamic: classifier
-Dynamic: description
-Dynamic: description-content-type
-Dynamic: home-page
-Dynamic: license-file
-Dynamic: provides-extra
-Dynamic: requires-dist
-Dynamic: requires-python
-Dynamic: summary
-
-# Opik Optimizer
-
-The Opik Opitmizer can refine your prompts to get better performance
-from your LLMs. You can use a variety of algorithms, including:
-
-* FewShotBayesianOptimizer
-* MiproOptimizer
-* MetaPromptOptimizer
-
-## Quickstart
-
-
-[Open Quickstart Notebook in Colab](https://colab.research.google.com/github/comet-ml/opik/blob/main/sdks/opik_optimizer/notebooks/OpikOptimizerIntro.ipynb)
-
-
-## Setup
-
-1. Configure Opik:
-   ```bash
-   # Install Comet ML CLI
-   pip install opik
-
-   # Configure your API key
-   opik configure
-   # When prompted, enter your Opik API key
-   ```
-
-2. Set up your environment variables:
-   ```bash
-   # OpenAI API key for LLM access
-   export OPENAI_API_KEY=your_openai_api_key
-   ```
-
-3. Install the package:
-   ```bash
-   pip install opik-optimizer
-   ```
-
-You'll need:
-
-1. An LLM model name
-2. An Opik Dataset (or Opik Dataset name)
-3. An Opik Metric (possibly a custom one)
-4. A starting prompt (string)
-
-## Example
-
-We have prepared some sample datasets for testing:
-
-* "tiny-test"
-* "halu-eval-300"
-* "hotpot-300"
-
-You can see how to use those below:
-
-```python
-from opik.evaluation.metrics import LevenshteinRatio
-from opik_optimizer import FewShotBayesianOptimizer
-from opik_optimizer.demo import get_or_create_dataset
-
-from opik_optimizer import (
-    MetricConfig,
-    TaskConfig,
-    from_dataset_field,
-    from_llm_response_text,
-)
-
-hot_pot_dataset = get_or_create_dataset("hotpot-300")
-
-# For chat prompts instruction doesn't need to contain input parameters from dataset examples.
-prompt_instruction = """
-Answer the question.
-"""
-project_name = "optimize-few-shot-bayesian-hotpot"
-
-optimizer = FewShotBayesianOptimizer(
-    model="gpt-4o-mini",
-    project_name=project_name,
-    min_examples=3,
-    max_examples=8,
-    n_threads=16,
-    seed=42,
-)
-
-metric_config = MetricConfig(
-    metric=LevenshteinRatio(project_name=project_name),
-    inputs={
-        "output": from_llm_response_text(),
-        "reference": from_dataset_field(name="answer"),
-    },
-)
-
-task_config = TaskConfig(
-    instruction_prompt=prompt_instruction,
-    input_dataset_fields=["question"],
-    output_dataset_field="answer",
-    use_chat_prompt=True,
-)
-
-result = optimizer.optimize_prompt(
-    dataset=hot_pot_dataset,
-    metric_config=metric_config,
-    task_config=task_config,
-    n_trials=10,
-    n_samples=150,
-)
-
-result.display()
-```
-
-More examples can be found in the `scripts` folder.
-
-## Installation
-
-```bash
-pip install opik-optimizer
-```
-
-## Development
-
-To use the Opik Optimizer from source:
-
-```bash
-git clone git clone git@github.com:comet-ml/opik
-cd sdks/opik_optimizer
-pip install -e .
-```
-
-## Requirements
-
-- Python 3.10+ < 3.13
-- Opik API key
-- OpenAI API key (or other LLM provider)

opik_optimizer-0.7.8.dist-info/RECORD

@@ -1,33 +0,0 @@
-opik_optimizer/__init__.py,sha256=Tbny_W_oUWKdf_33fM-nNTBqvQvarLfB8hMNd_goXT0,1245
-opik_optimizer/_throttle.py,sha256=ztub8qlwz4u0GVA2TIoLig0D1Cs0hJ7_o_SnT_C7Nmk,1360
-opik_optimizer/base_optimizer.py,sha256=f4gNX9j3Z3TGst8F0gm1nMHHpHKAlChmeCVAcTdTIR4,4883
-opik_optimizer/cache_config.py,sha256=EzF4RAzxhSG8vtMJANdiUpNHQ9HzL2CrCXp0iik0f4A,580
-opik_optimizer/logging_config.py,sha256=ELevhxtflYinTo-jVvyQYZbXG7FgAe_b5dPa9y5uLWw,2774
-opik_optimizer/meta_prompt_optimizer.py,sha256=tCCDHDjUEjpuicN-766HbwODZHeAzZGTXGqoRl9Npvc,46992
-opik_optimizer/optimization_result.py,sha256=9zdDV2MXeLYk7U8OqgMmSU-DdPV6qgYQWS2rtkO6Dzw,8693
-opik_optimizer/task_evaluator.py,sha256=MafDMaLeW0_yGPrumLvYF0HzQUKrnpAlM_0N_TPG8tw,3695
-opik_optimizer/utils.py,sha256=HivUsNzbt7BcuZeEvikdER1DaTPUFLJrpaVQ8raZYD8,3637
-opik_optimizer/data/hotpot-500.json,sha256=ToYC9l3lebrOoeWFhELgp63XVFYWcS0HODlQO9g5V3c,101861
-opik_optimizer/demo/__init__.py,sha256=KSpFYhzN7fTmLEsIaciRHwxcJDeAiX5NDmYLdPsfpT8,150
-opik_optimizer/demo/cache.py,sha256=5WqK8rSiijzU6s4VHIjLuL1LR5i1yHtY-x5FZTduSus,3669
-opik_optimizer/demo/datasets.py,sha256=-qmdeFk5X3CIcg9QX7nbGUAVo2khh0XGbmkHh8jgvSY,22633
-opik_optimizer/evolutionary_optimizer/__init__.py,sha256=OQ2ART5g-7EVGOISvTGY-AbmEyyDFEJJCsmJBzGJIpw,57
-opik_optimizer/evolutionary_optimizer/evolutionary_optimizer.py,sha256=652TAHa4U4C7_0_wBulDmizfVijqV9E3ngR38nLcpt4,76568
-opik_optimizer/few_shot_bayesian_optimizer/__init__.py,sha256=VuH7FOROyGcjMPryejtZC-5Y0QHlVTFLTGUDgNqRAFw,113
-opik_optimizer/few_shot_bayesian_optimizer/few_shot_bayesian_optimizer.py,sha256=nmf3mguehaIomiHPFnONvnxzJvvo15wMU-GLe_S0Gb0,15318
-opik_optimizer/few_shot_bayesian_optimizer/prompt_parameter.py,sha256=EDsSIFAUOfiZKWLrOAaBDB7Exk7cmIs4ccI95kVa7JY,3118
-opik_optimizer/few_shot_bayesian_optimizer/prompt_templates.py,sha256=HmvD-UeT3aKiiet5cUtULXe6iFPEOo6hxyDE0pH2LnQ,2424
-opik_optimizer/integrations/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-opik_optimizer/mipro_optimizer/__init__.py,sha256=CF9TVXjOxTobDO1kAS8CD4eyLVzEozxjfgoKwIO6ZpU,44
-opik_optimizer/mipro_optimizer/_lm.py,sha256=bcTy2Y5HjSaFQOATIpUaA86eIp3vKHaMuDI2_RvN2ww,16376
-opik_optimizer/mipro_optimizer/_mipro_optimizer_v2.py,sha256=CiQWe39LCp-81SZmLyMybIw2lc_0RBKcxclLEuSXQgI,39757
-opik_optimizer/mipro_optimizer/mipro_optimizer.py,sha256=WoBSFcgSdztwzNOkRRhxBW0aiR4B_ThxZRaebDgt_fA,15581
-opik_optimizer/mipro_optimizer/utils.py,sha256=wG1koygXfm_pvtA1jR-YaU4NATPbJZoTI7LE7l7df7g,3273
-opik_optimizer/optimization_config/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-opik_optimizer/optimization_config/configs.py,sha256=MYL9H2UAqeyGBlBGWbOZ-6Snto4ZMuXnypgvVuUSW1Y,1132
-opik_optimizer/optimization_config/mappers.py,sha256=RXgTMxPzTQ1AHGke6Zca6rTcfCI7IkCKhQYciaEGSAo,1698
-opik_optimizer-0.7.8.dist-info/licenses/LICENSE,sha256=dTRSwwCHdWeSjzodvnivYqcwi8x3Qfr21yv65QUWWBE,1062
-opik_optimizer-0.7.8.dist-info/METADATA,sha256=eYfFtFoEWgqLxVx2OYhmsXyHUOSjK6GArV8hvxdCgcs,3853
-opik_optimizer-0.7.8.dist-info/WHEEL,sha256=zaaOINJESkSfm_4HQVc5ssNzHCPXhJm0kEUakpsEHaU,91
-opik_optimizer-0.7.8.dist-info/top_level.txt,sha256=ondOlpq6_yFckqpxoAHSfzZS2N-JfgmA-QQhOJfz7m0,15
-opik_optimizer-0.7.8.dist-info/RECORD,,