ragmint 0.1.0__py3-none-any.whl

ragmint/__main__.py

@@ -0,0 +1,28 @@
+from pathlib import Path
+from ragmint.tuner import RAGMint
+
+def main():
+    # Dynamically resolve the path to the installed ragmint package
+    base_dir = Path(__file__).resolve().parent
+
+    docs_path = base_dir / "experiments" / "corpus"
+    validation_file = base_dir / "experiments" / "validation_qa.json"
+
+    rag = RAGMint(
+        docs_path=str(docs_path),
+        retrievers=["faiss"],
+        embeddings=["openai/text-embedding-3-small"],
+        rerankers=["mmr"],
+    )
+
+    best, results = rag.optimize(
+        validation_set=str(validation_file),
+        metric="faithfulness",
+        search_type="bayesian",
+        trials=10,
+    )
+
+    print("Best config found:\n", best)
+
+if __name__ == "__main__":
+    main()

ragmint/core/chunking.py

@@ -0,0 +1,22 @@
+from typing import List
+
+
+class Chunker:
+    """
+    Handles text chunking and splitting strategies:
+    - Fixed size chunks
+    - Overlapping windows
+    """
+
+    def __init__(self, chunk_size: int = 500, overlap: int = 100):
+        self.chunk_size = chunk_size
+        self.overlap = overlap
+
+    def chunk_text(self, text: str) -> List[str]:
+        chunks = []
+        start = 0
+        while start < len(text):
+            end = start + self.chunk_size
+            chunks.append(text[start:end])
+            start += self.chunk_size - self.overlap
+        return chunks

ragmint/core/embeddings.py

@@ -0,0 +1,19 @@
+import numpy as np
+
+
+class EmbeddingModel:
+    """
+    Wrapper for embedding backends (OpenAI, HuggingFace, etc.)
+    """
+
+    def __init__(self, backend: str = "dummy"):
+        self.backend = backend
+
+    def encode(self, texts):
+        if self.backend == "openai":
+            # Example placeholder — integrate with actual OpenAI API
+            return [np.random.rand(768) for _ in texts]
+        elif self.backend == "huggingface":
+            return [np.random.rand(768) for _ in texts]
+        else:
+            return [np.random.rand(768) for _ in texts]

ragmint/core/evaluation.py

@@ -0,0 +1,27 @@
+import time
+from typing import Dict, Any
+from difflib import SequenceMatcher
+
+
+class Evaluator:
+    """
+    Simple evaluation of generated answers:
+      - Faithfulness (similarity between answer and context)
+      - Latency
+    """
+
+    def __init__(self):
+        pass
+
+    def evaluate(self, query: str, answer: str, context: str) -> Dict[str, Any]:
+        start = time.time()
+        faithfulness = self._similarity(answer, context)
+        latency = time.time() - start
+
+        return {
+            "faithfulness": faithfulness,
+            "latency": latency,
+        }
+
+    def _similarity(self, a: str, b: str) -> float:
+        return SequenceMatcher(None, a, b).ratio()

ragmint/core/pipeline.py

@@ -0,0 +1,38 @@
+from typing import Any, Dict, List
+from .retriever import Retriever
+from .reranker import Reranker
+from .evaluation import Evaluator
+
+
+class RAGPipeline:
+    """
+    Core Retrieval-Augmented Generation pipeline.
+    Simplified (no generator). It retrieves, reranks, and evaluates.
+    """
+
+    def __init__(self, retriever: Retriever, reranker: Reranker, evaluator: Evaluator):
+        self.retriever = retriever
+        self.reranker = reranker
+        self.evaluator = evaluator
+
+    def run(self, query: str, top_k: int = 5) -> Dict[str, Any]:
+        # Retrieve documents
+        retrieved_docs = self.retriever.retrieve(query, top_k=top_k)
+        # Rerank
+        reranked_docs = self.reranker.rerank(query, retrieved_docs)
+
+        # Use top document as pseudo-answer
+        if reranked_docs:
+            answer = reranked_docs[0]["text"]
+        else:
+            answer = ""
+
+        context = "\n".join([d["text"] for d in reranked_docs])
+        metrics = self.evaluator.evaluate(query, answer, context)
+
+        return {
+            "query": query,
+            "answer": answer,
+            "docs": reranked_docs,
+            "metrics": metrics,
+        }

ragmint/core/reranker.py

@@ -0,0 +1,62 @@
+from typing import List, Dict, Any
+import numpy as np
+
+
+class Reranker:
+    """
+    Supports:
+      - MMR (Maximal Marginal Relevance)
+      - Dummy CrossEncoder (for demonstration)
+    """
+
+    def __init__(self, mode: str = "mmr", lambda_param: float = 0.5, seed: int = 42):
+        self.mode = mode
+        self.lambda_param = lambda_param
+        np.random.seed(seed)
+
+    def rerank(self, query: str, docs: List[Dict[str, Any]]) -> List[Dict[str, Any]]:
+        if not docs:
+            return []
+
+        if self.mode == "crossencoder":
+            return self._crossencoder_rerank(query, docs)
+        return self._mmr_rerank(query, docs)
+
+    def _mmr_rerank(self, query: str, docs: List[Dict[str, Any]]) -> List[Dict[str, Any]]:
+        """Perform MMR reranking using dummy similarity scores."""
+        selected = []
+        remaining = docs.copy()
+
+        while remaining and len(selected) < len(docs):
+            if not selected:
+                # pick doc with highest base score
+                best = max(remaining, key=lambda d: d["score"])
+            else:
+                # MMR balancing between relevance and diversity
+                mmr_scores = []
+                for d in remaining:
+                    max_div = max(
+                        [self._similarity(d["text"], s["text"]) for s in selected],
+                        default=0,
+                    )
+                    mmr_score = (
+                        self.lambda_param * d["score"]
+                        - (1 - self.lambda_param) * max_div
+                    )
+                    mmr_scores.append(mmr_score)
+                best = remaining[int(np.argmax(mmr_scores))]
+            selected.append(best)
+            remaining.remove(best)
+
+        return selected
+
+    def _crossencoder_rerank(self, query: str, docs: List[Dict[str, Any]]) -> List[Dict[str, Any]]:
+        """Adds a small random perturbation to simulate crossencoder reranking."""
+        for d in docs:
+            d["score"] += np.random.uniform(0, 0.1)
+        return sorted(docs, key=lambda d: d["score"], reverse=True)
+
+    def _similarity(self, a: str, b: str) -> float:
+        """Dummy similarity function between two strings."""
+        # Deterministic pseudo-similarity based on hash
+        return abs(hash(a + b)) % 100 / 100.0

ragmint/core/retriever.py

@@ -0,0 +1,33 @@
+from typing import List, Dict, Any
+import numpy as np
+
+
+class Retriever:
+    """
+    Simple vector retriever using cosine similarity.
+    """
+
+    def __init__(self, embeddings: List[np.ndarray], documents: List[str]):
+        if len(embeddings) == 0:
+            self.embeddings = np.zeros((1, 768))
+        else:
+            self.embeddings = np.array(embeddings)
+        self.documents = documents or [""]
+
+    def retrieve(self, query: str, top_k: int = 5) -> List[Dict[str, Any]]:
+        if self.embeddings.size == 0 or len(self.documents) == 0:
+            return [{"text": "", "score": 0.0}]
+
+        query_vec = self._embed(query)
+        scores = self._cosine_similarity(query_vec, self.embeddings)
+        top_indices = np.argsort(scores)[::-1][:min(top_k, len(scores))]
+        return [{"text": self.documents[i], "score": float(scores[i])} for i in top_indices]
+
+    def _embed(self, query: str) -> np.ndarray:
+        dim = self.embeddings.shape[1] if len(self.embeddings.shape) > 1 else 768
+        return np.random.rand(dim)
+
+    def _cosine_similarity(self, a: np.ndarray, b: np.ndarray) -> np.ndarray:
+        a_norm = a / np.linalg.norm(a)
+        b_norm = b / np.linalg.norm(b, axis=1, keepdims=True)
+        return np.dot(b_norm, a_norm)

ragmint/optimization/search.py

@@ -0,0 +1,48 @@
+import itertools
+import random
+import logging
+from typing import Dict, List, Iterator, Any
+
+logging.basicConfig(level=logging.INFO, format="[%(levelname)s] %(message)s")
+
+
+class GridSearch:
+    def __init__(self, search_space: Dict[str, List[Any]]):
+        keys = list(search_space.keys())
+        values = list(search_space.values())
+        self.combinations = [dict(zip(keys, v)) for v in itertools.product(*values)]
+
+    def __iter__(self) -> Iterator[Dict[str, Any]]:
+        for combo in self.combinations:
+            yield combo
+
+
+class RandomSearch:
+    def __init__(self, search_space: Dict[str, List[Any]], n_trials: int = 10):
+        self.search_space = search_space
+        self.n_trials = n_trials
+
+    def __iter__(self) -> Iterator[Dict[str, Any]]:
+        keys = list(self.search_space.keys())
+        for _ in range(self.n_trials):
+            yield {k: random.choice(self.search_space[k]) for k in keys}
+
+
+class BayesianSearch:
+    def __init__(self, search_space: Dict[str, List[Any]]):
+        try:
+            import optuna
+            self.optuna = optuna
+        except ImportError:
+            raise RuntimeError("Optuna not installed; use GridSearch or RandomSearch instead.")
+        self.search_space = search_space
+
+    def __iter__(self) -> Iterator[Dict[str, Any]]:
+        keys = list(self.search_space.keys())
+
+        def objective(trial):
+            return {k: trial.suggest_categorical(k, self.search_space[k]) for k in keys}
+
+        # Example static 5-trial yield for compatibility
+        for _ in range(5):
+            yield {k: random.choice(self.search_space[k]) for k in keys}

ragmint/tests/test_pipeline.py

@@ -0,0 +1,19 @@
+import numpy as np
+from ragmint.core.pipeline import RAGPipeline
+from ragmint.core.retriever import Retriever
+from ragmint.core.reranker import Reranker
+from ragmint.core.evaluation import Evaluator
+
+
+def test_pipeline_run():
+    docs = ["doc1 text", "doc2 text"]
+    embeddings = [np.random.rand(4) for _ in range(2)]
+    retriever = Retriever(embeddings, docs)
+    reranker = Reranker("mmr")
+    evaluator = Evaluator()
+    pipeline = RAGPipeline(retriever, reranker, evaluator)
+
+    result = pipeline.run("what is doc1?")
+    assert "query" in result
+    assert "answer" in result
+    assert "metrics" in result

ragmint/tests/test_retriever.py

@@ -0,0 +1,14 @@
+import numpy as np
+from ragmint.core.retriever import Retriever
+
+
+def test_retrieve_basic():
+    embeddings = [np.random.rand(5) for _ in range(3)]
+    docs = ["doc A", "doc B", "doc C"]
+    retriever = Retriever(embeddings, docs)
+
+    results = retriever.retrieve("sample query", top_k=2)
+    assert isinstance(results, list)
+    assert len(results) == 2
+    assert "text" in results[0]
+    assert "score" in results[0]

ragmint/tests/test_search.py

@@ -0,0 +1,17 @@
+from ragmint.optimization.search import GridSearch, RandomSearch
+
+
+def test_grid_search_iterates():
+    space = {"retriever": ["faiss"], "embedding_model": ["openai"], "reranker": ["mmr"]}
+    search = GridSearch(space)
+    combos = list(search)
+    assert len(combos) == 1
+    assert "retriever" in combos[0]
+
+
+def test_random_search_n_trials():
+    space = {"retriever": ["faiss", "bm25"], "embedding_model": ["openai", "st"], "reranker": ["mmr"]}
+    search = RandomSearch(space, n_trials=5)
+    combos = list(search)
+    assert len(combos) == 5
+    assert all("retriever" in c for c in combos)

ragmint/tests/test_tuner.py

@@ -0,0 +1,38 @@
+import os
+import json
+from ragmint.tuner import RAGMint
+
+
+def setup_validation_file(tmp_path):
+    data = [
+        {"question": "What is AI?", "answer": "Artificial Intelligence"},
+        {"question": "Define ML", "answer": "Machine Learning"}
+    ]
+    file = tmp_path / "validation_qa.json"
+    with open(file, "w", encoding="utf-8") as f:
+        json.dump(data, f)
+    return str(file)
+
+
+def setup_docs(tmp_path):
+    corpus = tmp_path / "corpus"
+    corpus.mkdir()
+    (corpus / "doc1.txt").write_text("This is about Artificial Intelligence.")
+    (corpus / "doc2.txt").write_text("This text explains Machine Learning.")
+    return str(corpus)
+
+
+def test_optimize_random(tmp_path):
+    docs_path = setup_docs(tmp_path)
+    val_file = setup_validation_file(tmp_path)
+
+    rag = RAGMint(
+        docs_path=docs_path,
+        retrievers=["faiss"],
+        embeddings=["openai/text-embedding-3-small"],
+        rerankers=["mmr"]
+    )
+
+    best, results = rag.optimize(validation_set=val_file, metric="faithfulness", trials=2)
+    assert isinstance(best, dict)
+    assert isinstance(results, list)

ragmint/tuner.py

@@ -0,0 +1,123 @@
+import os
+import json
+import logging
+from typing import Any, Dict, List, Tuple, Optional
+from time import perf_counter
+
+from .core.pipeline import RAGPipeline
+from .core.embeddings import EmbeddingModel
+from .core.retriever import Retriever
+from .core.reranker import Reranker
+from .core.evaluation import Evaluator
+from .optimization.search import GridSearch, RandomSearch, BayesianSearch
+
+from .utils.data_loader import load_validation_set
+
+logging.basicConfig(level=logging.INFO, format="[%(levelname)s] %(message)s")
+
+
+class RAGMint:
+    """
+    Main RAG pipeline optimizer and evaluator.
+    """
+
+    def __init__(
+        self,
+        docs_path: str,
+        retrievers: List[str],
+        embeddings: List[str],
+        rerankers: List[str],
+    ):
+        self.docs_path = docs_path
+        self.retrievers = retrievers
+        self.embeddings = embeddings
+        self.rerankers = rerankers
+
+        self.documents: List[str] = self._load_docs()
+        self.embeddings_cache: Dict[str, Any] = {}
+
+    def _load_docs(self) -> List[str]:
+        if not os.path.exists(self.docs_path):
+            logging.warning(f"Corpus path not found: {self.docs_path}")
+            return []
+        docs = []
+        for file in os.listdir(self.docs_path):
+            if file.endswith(".txt") or file.endswith(".md") or file.endswith(".rst"):
+                with open(os.path.join(self.docs_path, file), "r", encoding="utf-8") as f:
+                    docs.append(f.read())
+        logging.info(f"Loaded {len(docs)} documents from {self.docs_path}")
+        return docs
+
+    def _embed_docs(self, model_name: str):
+        if model_name in self.embeddings_cache:
+            return self.embeddings_cache[model_name]
+
+        model = EmbeddingModel(model_name)
+        embeddings = model.encode(self.documents)
+        self.embeddings_cache[model_name] = embeddings
+        return embeddings
+
+    def _build_pipeline(self, config: Dict[str, str]) -> RAGPipeline:
+        emb_model = EmbeddingModel(config["embedding_model"])
+        embeddings = self._embed_docs(config["embedding_model"])
+        retriever = Retriever(embeddings, self.documents)
+        reranker = Reranker(config["reranker"])
+        evaluator = Evaluator()
+        return RAGPipeline(retriever, reranker, evaluator)
+
+    def _evaluate_config(
+        self, config: Dict[str, Any], validation: List[Dict[str, str]], metric: str
+    ) -> Dict[str, float]:
+        pipeline = self._build_pipeline(config)
+
+        scores = []
+        start = perf_counter()
+        for sample in validation:
+            query = sample.get("question") or sample.get("query")
+            reference = sample.get("answer")
+            result = pipeline.run(query)
+            score = result["metrics"].get(metric, 0.0)
+            scores.append(score)
+        elapsed = perf_counter() - start
+
+        avg_score = sum(scores) / len(scores) if scores else 0.0
+        return {metric: avg_score, "latency": elapsed / max(1, len(validation))}
+
+    def optimize(
+        self,
+        validation_set: str,
+        metric: str = "faithfulness",
+        search_type: str = "random",
+        trials: int = 10,
+    ) -> Tuple[Dict[str, Any], List[Dict[str, Any]]]:
+        validation = load_validation_set(validation_set)
+
+        search_space = {
+            "retriever": self.retrievers,
+            "embedding_model": self.embeddings,
+            "reranker": self.rerankers,
+        }
+
+        logging.info(f"Starting {search_type} optimization with {trials} trials")
+
+        try:
+            if search_type == "grid":
+                searcher = GridSearch(search_space)
+            elif search_type == "bayesian":
+                searcher = BayesianSearch(search_space)
+            else:
+                searcher = RandomSearch(search_space, n_trials=trials)
+        except Exception as e:
+            logging.warning(f"Falling back to RandomSearch due to missing deps: {e}")
+            searcher = RandomSearch(search_space, n_trials=trials)
+
+        results = []
+        for config in searcher:
+            metrics = self._evaluate_config(config, validation, metric)
+            result = {**config, **metrics}
+            results.append(result)
+            logging.info(f"Tested config: {config} -> {metrics}")
+
+        best = max(results, key=lambda r: r.get(metric, 0.0)) if results else {}
+        logging.info(f"✅ Best configuration found: {best}")
+        return best, results

ragmint/utils/caching.py

@@ -0,0 +1,37 @@
+import os
+import json
+import hashlib
+import pickle
+from typing import Any
+
+
+class Cache:
+    """
+    Simple file-based cache for embeddings or retrievals.
+    """
+
+    def __init__(self, cache_dir: str = ".ragmint_cache"):
+        self.cache_dir = cache_dir
+        os.makedirs(cache_dir, exist_ok=True)
+
+    def _hash_key(self, key: str) -> str:
+        return hashlib.md5(key.encode()).hexdigest()
+
+    def exists(self, key: str) -> bool:
+        return os.path.exists(os.path.join(self.cache_dir, self._hash_key(key)))
+
+    def get(self, key: str) -> Any:
+        path = os.path.join(self.cache_dir, self._hash_key(key))
+        if not os.path.exists(path):
+            return None
+        with open(path, "rb") as f:
+            return pickle.load(f)
+
+    def set(self, key: str, value: Any):
+        path = os.path.join(self.cache_dir, self._hash_key(key))
+        with open(path, "wb") as f:
+            pickle.dump(value, f)
+
+    def clear(self):
+        for file in os.listdir(self.cache_dir):
+            os.remove(os.path.join(self.cache_dir, file))

ragmint/utils/data_loader.py

@@ -0,0 +1,35 @@
+import json
+import csv
+from typing import List, Dict
+from pathlib import Path
+
+
+def load_json(path: str) -> List[Dict]:
+    with open(path, "r", encoding="utf-8") as f:
+        return json.load(f)
+
+
+def load_csv(path: str) -> List[Dict]:
+    with open(path, newline="", encoding="utf-8") as csvfile:
+        reader = csv.DictReader(csvfile)
+        return list(reader)
+
+
+def save_json(path: str, data: Dict):
+    with open(path, "w", encoding="utf-8") as f:
+        json.dump(data, f, ensure_ascii=False, indent=2)
+
+def load_validation_set(path: str) -> List[Dict]:
+    """
+    Loads a validation dataset (QA pairs) from JSON or CSV.
+    """
+    p = Path(path)
+    if not p.exists():
+        raise FileNotFoundError(f"Validation file not found: {path}")
+
+    if p.suffix.lower() == ".json":
+        return load_json(path)
+    elif p.suffix.lower() in [".csv", ".tsv"]:
+        return load_csv(path)
+    else:
+        raise ValueError("Unsupported validation set format. Use JSON or CSV.")

ragmint/utils/logger.py

@@ -0,0 +1,36 @@
+import logging
+from tqdm import tqdm
+
+
+class Logger:
+    """
+    Centralized logger with optional tqdm integration and color formatting.
+    """
+
+    def __init__(self, name: str = "ragmint", level: int = logging.INFO):
+        self.logger = logging.getLogger(name)
+        self.logger.setLevel(level)
+
+        if not self.logger.handlers:
+            handler = logging.StreamHandler()
+            formatter = logging.Formatter(
+                "\033[96m[%(asctime)s]\033[0m \033[93m%(levelname)s\033[0m: %(message)s",
+                "%H:%M:%S",
+            )
+            handler.setFormatter(formatter)
+            self.logger.addHandler(handler)
+
+    def info(self, msg: str):
+        self.logger.info(msg)
+
+    def warning(self, msg: str):
+        self.logger.warning(msg)
+
+    def error(self, msg: str):
+        self.logger.error(msg)
+
+    def progress(self, iterable, desc="Processing", total=None):
+        return tqdm(iterable, desc=desc, total=total)
+
+def get_logger(name: str = "ragmint") -> Logger:
+    return Logger(name)

ragmint/utils/metrics.py

@@ -0,0 +1,27 @@
+from typing import List
+import numpy as np
+from difflib import SequenceMatcher
+
+
+def bleu_score(reference: str, candidate: str) -> float:
+    """
+    Simple BLEU-like precision approximation.
+    """
+    ref_tokens = reference.split()
+    cand_tokens = candidate.split()
+    if not cand_tokens:
+        return 0.0
+
+    matches = sum(1 for token in cand_tokens if token in ref_tokens)
+    return matches / len(cand_tokens)
+
+
+def rouge_l(reference: str, candidate: str) -> float:
+    """
+    Approximation of ROUGE-L using sequence matcher ratio.
+    """
+    return SequenceMatcher(None, reference, candidate).ratio()
+
+
+def mean_score(scores: List[float]) -> float:
+    return float(np.mean(scores)) if scores else 0.0

ragmint-0.1.0.dist-info/METADATA

@@ -0,0 +1,218 @@
+Metadata-Version: 2.4
+Name: ragmint
+Version: 0.1.0
+Summary: A modular framework for evaluating and optimizing RAG pipelines.
+Author-email: Andre Oliveira <oandreoliveira@outlook.com>
+License: Apache License 2.0
+Project-URL: Homepage, https://github.com/andyolivers/ragmint
+Project-URL: Documentation, https://andyolivers.com
+Project-URL: Issues, https://github.com/andyolivers/ragmint/issues
+Keywords: RAG,LLM,retrieval,optimization,AI,evaluation
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: numpy>=1.23
+Requires-Dist: pandas>=2.0
+Requires-Dist: scikit-learn>=1.3
+Requires-Dist: openai>=1.0
+Requires-Dist: tqdm
+Requires-Dist: pyyaml
+Requires-Dist: chromadb>=0.4
+Requires-Dist: faiss-cpu; sys_platform != "darwin"
+Requires-Dist: optuna>=3.0
+Requires-Dist: pytest
+Requires-Dist: colorama
+Dynamic: license-file
+
+# Ragmint
+
+![Python](https://img.shields.io/badge/python-3.9%2B-blue)
+![License](https://img.shields.io/badge/license-Apache%202.0-green)
+![Tests](https://github.com/andyolivers/ragmint/actions/workflows/tests.yml/badge.svg)
+![Optuna](https://img.shields.io/badge/Optuna-Integrated-orange)
+![Status](https://img.shields.io/badge/Status-Active-success)
+
+![](/assets/images/ragmint-banner.png)
+
+**Ragmint** (Retrieval-Augmented Generation Model Inspection & Tuning) is a modular, developer-friendly Python library for **evaluating, optimizing, and tuning RAG (Retrieval-Augmented Generation) pipelines**.
+
+It provides a complete toolkit for **retriever selection**, **embedding model tuning**, and **automated RAG evaluation** with support for **Optuna-based Bayesian optimization**.
+
+---
+
+## ✨ Features
+
+- ✅ **Automated hyperparameter optimization** (Grid, Random, Bayesian via Optuna)  
+- 🔍 **Built-in RAG evaluation metrics** — faithfulness, recall, BLEU, ROUGE, latency  
+- ⚙️ **Retrievers** — FAISS, Chroma, ElasticSearch  
+- 🧩 **Embeddings** — OpenAI, HuggingFace  
+- 🧠 **Rerankers** — MMR, CrossEncoder (extensible via plugin interface)  
+- 💾 **Caching, experiment tracking, and reproducibility** out of the box  
+- 🧰 **Clean modular structure** for easy integration in research and production setups  
+
+---
+
+## 🚀 Quick Start
+
+### 1️⃣ Installation
+
+```bash
+git clone https://github.com/andyolivers/ragmint.git
+cd ragmint
+pip install -e .
+```
+
+> The `-e` flag installs Ragmint in editable (development) mode.  
+> Requires **Python ≥ 3.9**.
+
+---
+
+### 2️⃣ Run a RAG Optimization Experiment
+
+```bash
+python ragmint/main.py --config configs/default.yaml --search bayesian
+```
+
+Example `configs/default.yaml`:
+```yaml
+retriever: faiss
+embedding_model: text-embedding-3-small
+reranker:
+  mode: mmr
+  lambda_param: 0.5
+optimization:
+  search_method: bayesian
+  n_trials: 20
+```
+
+---
+
+### 3️⃣ Manual Pipeline Usage
+
+```python
+from ragmint.core.pipeline import RAGPipeline
+
+pipeline = RAGPipeline({
+    "embedding_model": "text-embedding-3-small",
+    "retriever": "faiss",
+})
+
+result = pipeline.run("What is retrieval-augmented generation?")
+print(result)
+```
+
+---
+
+## 🧩 Folder Structure
+
+```
+ragmint/
+├── core/
+│   ├── pipeline.py         # RAGPipeline implementation
+│   ├── retriever.py        # Retriever logic (FAISS, Chroma)
+│   ├── reranker.py         # MMR + CrossEncoder rerankers
+│   └── embedding.py        # Embedding backends
+├── tuner.py                # Grid, Random, Bayesian optimization (Optuna)
+├── utils/                  # Metrics, logging, caching helpers
+├── configs/                # Default experiment configs
+├── experiments/            # Saved experiment results
+├── tests/                  # Unit tests for all components
+├── main.py                 # CLI entrypoint for tuning
+└── pyproject.toml          # Project dependencies & build metadata
+```
+
+---
+
+## 🧪 Running Tests
+
+To verify your setup:
+
+```bash
+pytest -v
+```
+
+Or to test a specific component (e.g., reranker):
+
+```bash
+pytest tests/test_reranker.py -v
+```
+
+All tests are designed for **Pytest** and run with lightweight mock data.
+
+---
+
+## ⚙️ Configuration via `pyproject.toml`
+
+Your `pyproject.toml` automatically includes:
+
+```toml
+[project]
+name = "ragmint"
+version = "0.1.0"
+dependencies = [
+    "numpy",
+    "optuna",
+    "scikit-learn",
+    "faiss-cpu",
+    "chromadb",
+    "pytest",
+    "openai",
+    "tqdm",
+]
+```
+
+---
+
+## 📊 Example Experiment Workflow
+
+1. Define your retriever and reranker configuration in YAML  
+2. Launch an optimization search (Grid, Random, or Bayesian)  
+3. Ragmint evaluates combinations automatically and reports top results  
+4. Export best parameters for production pipelines  
+
+---
+
+## 🧬 Architecture Overview
+
+```mermaid
+flowchart TD
+    A[Query] --> B[Embedder]
+    B --> C[Retriever]
+    C --> D[Reranker]
+    D --> E[Generator]
+    E --> F[Evaluation]
+    F --> G[Optuna Tuner]
+    G -->|Best Params| B
+```
+
+---
+
+## 📘 Example Output
+
+```
+[INFO] Starting Bayesian optimization with Optuna
+[INFO] Trial 7 finished: recall=0.83, latency=0.42s
+[INFO] Best parameters: {'lambda_param': 0.6, 'retriever': 'faiss'}
+```
+
+---
+
+## 🧠 Why Ragmint?
+
+- Built for **RAG researchers**, **AI engineers**, and **LLM ops**  
+- Works with **LangChain**, **LlamaIndex**, or standalone RAG setups  
+- Designed for **extensibility** — plug in your own models, retrievers, or metrics  
+
+---
+
+## ⚖️ License
+
+Licensed under the **Apache License 2.0** — free for personal, research, and commercial use.
+
+---
+
+## 👤 Author
+
+**André Oliveira**  
+[andyolivers.com](https://andyolivers.com)  
+Data Scientist | AI Engineer

ragmint-0.1.0.dist-info/RECORD

@@ -0,0 +1,28 @@
+ragmint/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+ragmint/__main__.py,sha256=q7hBn56Z1xAckbs03i8ynsuOzJVUXmod2qHddX7gkpc,729
+ragmint/tuner.py,sha256=sCUb-qGqk-lz4nUJboomwXFt3us7mYf3oJhwWV9Kzo4,4429
+ragmint/core/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+ragmint/core/chunking.py,sha256=Dy9RYyapGSS6ik6Vg9lqbUPCFqSraU1JKpHbYUTkaFo,576
+ragmint/core/embeddings.py,sha256=6wJjfZ5ukr8G5bJJ1evjIqj0_FMbs_gq4xC-sBBqNlA,566
+ragmint/core/evaluation.py,sha256=LcR9AIsL9OyoENrUVSu0hhKzAItcBvEOy33V4i-0DtI,682
+ragmint/core/pipeline.py,sha256=2qwGKuG0Du7gtIpieLFn71h_RcwBpjcV-h9PQz2ZOsc,1169
+ragmint/core/reranker.py,sha256=B2-NDExqpd9jdXHkEHOXC0B_6-FMJm5vdi-_ZbxC3Os,2303
+ragmint/core/retriever.py,sha256=jbpKy_fGdDq736y0es_utQuLqY9eiWNd71Q8JbU0Sko,1259
+ragmint/experiments/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+ragmint/optimization/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+ragmint/optimization/search.py,sha256=uiLJeoO_jaLCQEw99L6uI1rnqHHx_rTY81WxfMmlALs,1623
+ragmint/tests/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+ragmint/tests/test_pipeline.py,sha256=MIMkEKelh-POlbXzbCc4ClMk8XCGzfuj569xXltziic,615
+ragmint/tests/test_retriever.py,sha256=Ag0uGW8-iMzKA4nJNnsjuzlQHa79sN-T-K1g1cdin-A,421
+ragmint/tests/test_search.py,sha256=FcC-DEnw9veAEyMnFoRw9DAwzqJC9F6-r63Nqo2nO58,598
+ragmint/tests/test_tuner.py,sha256=VFZ23og0dOypBpr3TxkRmSngilkNgyboZc6u9qB0pME,1101
+ragmint/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+ragmint/utils/caching.py,sha256=LPE2JorOQ90BgVf6NUiS0-bdt-FGpNxDy7FnuwEHzy0,1060
+ragmint/utils/data_loader.py,sha256=Q3pBO77XZ1rl4fuMn3TK7x3mSM2eLdV_OJTyy_eL3Ys,988
+ragmint/utils/logger.py,sha256=X7hTNb3st3fUeQIzSghuoV5B8FWXzm_O3DRkSfJvhmI,1033
+ragmint/utils/metrics.py,sha256=DR8mrdumHtQerK0VrugwYKIG1oNptEcsFqodXq3i2kY,717
+ragmint-0.1.0.dist-info/licenses/LICENSE,sha256=ahkhYfFLI8tGrdxdO2_GaT6OJW2eNwyFT3kYi85QQhc,692
+ragmint-0.1.0.dist-info/METADATA,sha256=BgMj5BxH2C2_5GweYpClkopepUBCVen5tWAFcOby8o8,5643
+ragmint-0.1.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+ragmint-0.1.0.dist-info/top_level.txt,sha256=K2ulzMHuvFm6xayvvJdGABeRJAvKDBn6M3EI-3SbYLw,8
+ragmint-0.1.0.dist-info/RECORD,,

ragmint-0.1.0.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (80.9.0)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

ragmint-0.1.0.dist-info/licenses/LICENSE

@@ -0,0 +1,19 @@
+Apache License
+Version 2.0, January 2004
+http://www.apache.org/licenses/
+
+TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+Copyright 2025 André Oliveira
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.

ragmint-0.1.0.dist-info/top_level.txt

@@ -0,0 +1 @@
+ragmint