File size: 8,707 Bytes
1ffeec5 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 | """
BeastBullet + RAG: Personal Knowledge Assistant
Query your documents, emails, notes - 100% local
"""
import os
from pathlib import Path
from typing import List, Dict
import json
class DocumentRAG:
"""Document retrieval for personal knowledge."""
def __init__(self, vector_db="chromadb", collection_name="my_knowledge"):
self.collection_name = collection_name
self.db = self._init_vector_db(vector_db)
def _init_vector_db(self, db_type: str):
"""Initialize vector database."""
if db_type == "chromadb":
try:
import chromadb
client = chromadb.Client()
collection = client.get_or_create_collection(
name=self.collection_name
)
print(f"✅ ChromaDB initialized: {self.collection_name}")
return collection
except ImportError:
print("⚠️ Install chromadb: pip install chromadb")
return None
else:
raise ValueError(f"Unsupported vector DB: {db_type}")
def index_directory(self, directory: str, extensions: List[str] = None):
"""Index all documents in a directory."""
if not self.db:
return
if extensions is None:
extensions = ['.txt', '.md', '.pdf', '.docx', '.py', '.js', '.json']
directory = Path(directory).expanduser()
print(f"\n📁 Indexing: {directory}")
indexed = 0
for ext in extensions:
for file_path in directory.rglob(f"*{ext}"):
try:
# Read file
if ext == '.pdf':
content = self._read_pdf(file_path)
else:
with open(file_path, 'r', encoding='utf-8', errors='ignore') as f:
content = f.read()
if not content.strip():
continue
# Chunk content
chunks = self._chunk_text(content)
# Add to vector DB
for i, chunk in enumerate(chunks):
self.db.add(
documents=[chunk],
ids=[f"{file_path}_{i}"],
metadatas=[{
"source": str(file_path),
"chunk": i,
"total_chunks": len(chunks)
}]
)
indexed += 1
print(f" ✅ {file_path.name} ({len(chunks)} chunks)")
except Exception as e:
print(f" ⚠️ Error indexing {file_path.name}: {e}")
print(f"\n✅ Indexed {indexed} documents")
def _chunk_text(self, text: str, chunk_size: int = 500) -> List[str]:
"""Split text into chunks."""
words = text.split()
chunks = []
for i in range(0, len(words), chunk_size):
chunk = ' '.join(words[i:i + chunk_size])
if chunk.strip():
chunks.append(chunk)
return chunks
def _read_pdf(self, file_path: Path) -> str:
"""Read PDF file."""
try:
import PyPDF2
with open(file_path, 'rb') as f:
reader = PyPDF2.PdfReader(f)
text = ""
for page in reader.pages:
text += page.extract_text()
return text
except ImportError:
print("⚠️ Install PyPDF2: pip install PyPDF2")
return ""
def retrieve(self, query: str, n_results: int = 5) -> Dict:
"""Retrieve relevant documents."""
if not self.db:
return {"documents": [], "sources": []}
results = self.db.query(
query_texts=[query],
n_results=n_results
)
# Format results
documents = results['documents'][0] if results['documents'] else []
metadatas = results['metadatas'][0] if results['metadatas'] else []
sources = []
for meta in metadatas:
if meta['source'] not in [s['file'] for s in sources]:
sources.append({
"file": meta['source'],
"chunks": 1
})
else:
for s in sources:
if s['file'] == meta['source']:
s['chunks'] += 1
return {
"documents": documents,
"sources": sources,
"context": "\n\n".join(documents)
}
class PersonalKnowledgeAssistant:
"""Personal knowledge assistant with BeastBullet + RAG."""
def __init__(self):
from code.integration_example import TinyLlamaIntegration, ExpertLoader
print("🧠 Loading BeastBullet...")
loader = ExpertLoader("experts/")
loader.load_encoder()
loader.load_all_experts()
self.bb = TinyLlamaIntegration(
experts=loader.experts,
encoder=loader.encoder
)
self.rag = DocumentRAG()
print("✅ Personal Knowledge Assistant ready!\n")
def index_my_knowledge(self):
"""Index common personal knowledge locations."""
print("📚 Indexing your knowledge base...")
# Common locations
locations = [
"~/Documents",
"~/Downloads",
"~/Desktop",
"~/Notes",
]
for location in locations:
path = Path(location).expanduser()
if path.exists():
self.rag.index_directory(path)
def query(self, question: str) -> Dict:
"""Query your personal knowledge."""
print("=" * 70)
print(f"Question: {question}")
print("=" * 70)
# Retrieve relevant documents
retrieval = self.rag.retrieve(question)
if not retrieval["documents"]:
print("\n⚠️ No relevant documents found in your knowledge base")
result = self.bb.query(question)
return result
# Build context-aware query
context_query = f"""Based on my personal documents, answer: {question}
My Documents:
{retrieval['context']}
Provide a clear answer based on my documents."""
# Query BeastBullet with context
result = self.bb.query(context_query)
result["sources"] = retrieval["sources"]
return result
def display_result(self, result: Dict):
"""Display result."""
print("\n" + "=" * 70)
print("🧠 Answer:")
print("=" * 70)
print(result.get("answer", "No answer generated"))
if "sources" in result and result["sources"]:
print("\n" + "=" * 70)
print("📁 Sources from your documents:")
print("=" * 70)
for source in result["sources"]:
print(f" • {source['file']} ({source['chunks']} relevant sections)")
print("\n" + "=" * 70)
print(f"Experts: {', '.join(result.get('experts_used', []))}")
print(f"Confidence: {result.get('confidence', 0):.1%}")
print("=" * 70)
# ============================================================================
# DEMO
# ============================================================================
def setup():
"""Setup and index documents."""
assistant = PersonalKnowledgeAssistant()
assistant.index_my_knowledge()
return assistant
def interactive():
"""Interactive mode."""
assistant = setup()
print("\n🧠 Personal Knowledge Assistant - Interactive Mode")
print("=" * 70)
print("Ask questions about your documents. Type 'exit' to quit.")
print("=" * 70)
print()
while True:
try:
question = input("You: ").strip()
if question.lower() in ['exit', 'quit', 'q']:
print("\n👋 Goodbye!")
break
if not question:
continue
result = assistant.query(question)
assistant.display_result(result)
print("\n")
except KeyboardInterrupt:
print("\n\n👋 Goodbye!")
break
if __name__ == "__main__":
interactive()
|