File size: 8,751 Bytes
793d027 2cec50c 793d027 408e7e3 793d027 408e7e3 2cec50c 793d027 408e7e3 793d027 408e7e3 793d027 408e7e3 793d027 408e7e3 793d027 408e7e3 793d027 408e7e3 793d027 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 | """ChromaDB vector store for unstructured document RAG."""
from pathlib import Path
from typing import Optional
import hashlib
from ..config import get_settings
# Project paths
PROJECT_ROOT = Path(__file__).parent.parent.parent
DOCS_DIR = PROJECT_ROOT / "docs"
def get_chroma_client():
"""Get ChromaDB persistent client."""
import chromadb
chroma_dir = get_settings().chroma_db_dir
chroma_dir.mkdir(parents=True, exist_ok=True)
return chromadb.PersistentClient(path=str(chroma_dir))
def get_embedding_function():
"""Get the embedding function for ChromaDB."""
from chromadb.utils import embedding_functions
return embedding_functions.SentenceTransformerEmbeddingFunction(
model_name="all-MiniLM-L6-v2"
)
def extract_pdf_text(pdf_path: Path) -> str:
"""Extract text from PDF file."""
from pypdf import PdfReader
reader = PdfReader(pdf_path)
text = ""
for page in reader.pages:
text += page.extract_text() + "\n\n"
return text
def chunk_text(text: str, chunk_size: int = 1000, chunk_overlap: int = 200) -> list[str]:
"""Split text into chunks for embedding."""
from langchain_text_splitters import RecursiveCharacterTextSplitter
splitter = RecursiveCharacterTextSplitter(
chunk_size=chunk_size,
chunk_overlap=chunk_overlap,
separators=["\n\n", "\n", ". ", " ", ""]
)
return splitter.split_text(text)
def generate_doc_id(text: str, index: int) -> str:
"""Generate a unique document ID."""
hash_input = f"{text[:100]}_{index}"
return hashlib.md5(hash_input.encode()).hexdigest()
def init_idsa_guidelines_collection():
"""Initialize the IDSA treatment guidelines collection."""
client = get_chroma_client()
ef = get_embedding_function()
# Delete existing collection if exists
try:
client.delete_collection("idsa_treatment_guidelines")
except Exception:
pass
collection = client.create_collection(
name="idsa_treatment_guidelines",
embedding_function=ef,
metadata={
"source": "IDSA 2024 Guidance",
"doi": "10.1093/cid/ciae403",
"description": "Antimicrobial-Resistant Gram-Negative Infections Treatment Guidelines"
}
)
return collection
def init_mic_reference_collection():
"""Initialize the MIC reference documentation collection."""
client = get_chroma_client()
ef = get_embedding_function()
# Delete existing collection if exists
try:
client.delete_collection("mic_reference_docs")
except Exception:
pass
collection = client.create_collection(
name="mic_reference_docs",
embedding_function=ef,
metadata={
"source": "EUCAST Breakpoint Tables",
"version": "16.0",
"description": "MIC Breakpoint Reference Documentation"
}
)
return collection
def classify_chunk_pathogen(text: str) -> str:
"""Classify which pathogen type a chunk relates to."""
text_lower = text.lower()
pathogen_keywords = {
"ESBL-E": ["esbl", "extended-spectrum beta-lactamase", "esbl-e", "esbl-producing"],
"CRE": ["carbapenem-resistant enterobacterales", "cre", "carbapenemase"],
"CRAB": ["acinetobacter baumannii", "crab", "carbapenem-resistant acinetobacter"],
"DTR-PA": ["pseudomonas aeruginosa", "dtr-p", "difficult-to-treat resistance"],
"S.maltophilia": ["stenotrophomonas maltophilia", "s. maltophilia"],
"AmpC-E": ["ampc", "ampc-e", "ampc-producing"],
}
for pathogen, keywords in pathogen_keywords.items():
for keyword in keywords:
if keyword in text_lower:
return pathogen
return "General"
def import_idsa_guidelines() -> int:
"""Import IDSA guidelines PDF into ChromaDB."""
print("Importing IDSA guidelines into ChromaDB...")
pdf_path = DOCS_DIR / "antibiotic_guidelines" / "ciae403.pdf"
if not pdf_path.exists():
print(f" Warning: {pdf_path} not found, skipping...")
return 0
# Extract text from PDF
print(" Extracting text from PDF...")
text = extract_pdf_text(pdf_path)
# Chunk the text
print(" Chunking text...")
chunks = chunk_text(text)
# Initialize collection
collection = init_idsa_guidelines_collection()
# Prepare documents for insertion
documents = []
metadatas = []
ids = []
for i, chunk in enumerate(chunks):
documents.append(chunk)
metadatas.append({
"source": "ciae403.pdf",
"chunk_index": i,
"pathogen_type": classify_chunk_pathogen(chunk),
"page_estimate": i // 3 # Rough estimate
})
ids.append(generate_doc_id(chunk, i))
# Add to collection
print(f" Adding {len(documents)} chunks to collection...")
collection.add(
documents=documents,
metadatas=metadatas,
ids=ids
)
print(f" Imported {len(documents)} chunks from IDSA guidelines")
return len(documents)
def import_mic_reference() -> int:
"""Import MIC breakpoint PDF into ChromaDB."""
print("Importing MIC reference PDF into ChromaDB...")
pdf_path = DOCS_DIR / "mic_breakpoints" / "v_16.0_Breakpoint_Tables.pdf"
if not pdf_path.exists():
print(f" Warning: {pdf_path} not found, skipping...")
return 0
# Extract text from PDF
print(" Extracting text from PDF...")
text = extract_pdf_text(pdf_path)
# Chunk the text
print(" Chunking text...")
chunks = chunk_text(text, chunk_size=800, chunk_overlap=150)
# Initialize collection
collection = init_mic_reference_collection()
# Prepare documents for insertion
documents = []
metadatas = []
ids = []
for i, chunk in enumerate(chunks):
documents.append(chunk)
metadatas.append({
"source": "v_16.0_Breakpoint_Tables.pdf",
"chunk_index": i,
"document_type": "mic_reference"
})
ids.append(generate_doc_id(chunk, i))
# Add to collection
print(f" Adding {len(documents)} chunks to collection...")
collection.add(
documents=documents,
metadatas=metadatas,
ids=ids
)
print(f" Imported {len(documents)} chunks from MIC reference")
return len(documents)
def get_collection(name: str) -> Optional[object]:
"""Get a collection by name."""
client = get_chroma_client()
ef = get_embedding_function()
try:
return client.get_collection(name=name, embedding_function=ef)
except Exception:
return None
def search_guidelines(
query: str,
n_results: int = 5,
pathogen_filter: str = None
) -> list[dict]:
"""Search the IDSA guidelines collection."""
collection = get_collection("idsa_treatment_guidelines")
if collection is None:
return []
where_filter = None
if pathogen_filter:
where_filter = {"pathogen_type": pathogen_filter}
results = collection.query(
query_texts=[query],
n_results=n_results,
where=where_filter,
include=["documents", "metadatas", "distances"]
)
# Format results
formatted = []
for i in range(len(results['documents'][0])):
formatted.append({
"content": results['documents'][0][i],
"metadata": results['metadatas'][0][i],
"distance": results['distances'][0][i]
})
return formatted
def search_mic_reference(query: str, n_results: int = 3) -> list[dict]:
"""Search the MIC reference collection."""
collection = get_collection("mic_reference_docs")
if collection is None:
return []
results = collection.query(
query_texts=[query],
n_results=n_results,
include=["documents", "metadatas", "distances"]
)
# Format results
formatted = []
for i in range(len(results['documents'][0])):
formatted.append({
"content": results['documents'][0][i],
"metadata": results['metadatas'][0][i],
"distance": results['distances'][0][i]
})
return formatted
def import_all_vectors() -> dict:
"""Import all PDFs into ChromaDB."""
print(f"\n{'='*50}")
print("ChromaDB Vector Import")
print(f"{'='*50}\n")
results = {
"idsa_guidelines": import_idsa_guidelines(),
"mic_reference": import_mic_reference(),
}
print(f"\n{'='*50}")
print("Vector Import Summary:")
for collection, count in results.items():
print(f" {collection}: {count} chunks")
print(f"{'='*50}\n")
return results
if __name__ == "__main__":
import_all_vectors()
|