Spaces:
Sleeping
Sleeping
File size: 7,757 Bytes
402e33f |
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 |
# # from langchain_chroma import Chroma
# # from langchain_openai import OpenAIEmbeddings
# from langchain_community.embeddings import HuggingFaceEmbeddings
# from utils.vector_store import get_vector_store
# import os
# # Define different embedding model options
# # EMBEDDING_CONFIGS = {
# # "Accuracy (OpenAI text-embedding-3-large)": OpenAIEmbeddings(model="text-embedding-3-large"),
# # "Performance (OpenAI text-embedding-3-small)": OpenAIEmbeddings(model="text-embedding-3-small"),
# # "Instruction-based (HuggingFace bge-large-en)": HuggingFaceEmbeddings(model_name="BAAI/bge-large-en"),
# # "QA Optimized (HuggingFace e5-large-v2)": HuggingFaceEmbeddings(model_name="intfloat/e5-large-v2"),
# # }
# EMBEDDING_CONFIGS = {
# "General-purpose (bge-large-en)": HuggingFaceEmbeddings(
# model_name="BAAI/bge-large-en",
# model_kwargs={"device": "cpu"}, # or "cuda" if you have GPU
# encode_kwargs={"normalize_embeddings": True}
# ),
# "Fast & lightweight (bge-small-en)": HuggingFaceEmbeddings(
# model_name="BAAI/bge-small-en",
# model_kwargs={"device": "cpu"},
# encode_kwargs={"normalize_embeddings": True}
# ),
# "QA optimized (e5-large-v2)": HuggingFaceEmbeddings(
# model_name="intfloat/e5-large-v2",
# model_kwargs={"device": "cpu"},
# encode_kwargs={"normalize_embeddings": True}
# ),
# "Instruction-tuned (instructor-large)": HuggingFaceEmbeddings(
# model_name="hkunlp/instructor-large",
# model_kwargs={"device": "cpu"},
# encode_kwargs={"normalize_embeddings": True}
# ),
# }
# # Default vector store path
# VECTOR_STORE_PATH = "./vector_stores/mes_db"
# def test_retriever_with_embeddings(query: str, embedding_model, k: int = 3):
# """Retrieve documents using a specific embedding model."""
# vector_store = get_vector_store(
# persist_directory=VECTOR_STORE_PATH,
# embedding=embedding_model
# )
# retriever = vector_store.as_retriever(search_kwargs={"k": k})
# docs = retriever.get_relevant_documents(query)
# # Deduplicate based on page_content
# seen = set()
# unique_docs = []
# for doc in docs:
# if doc.page_content not in seen:
# seen.add(doc.page_content)
# unique_docs.append(doc)
# return unique_docs
# def compare_embeddings(query: str, k: int = 3):
# print(f"\n=== Comparing embeddings for: '{query}' ===\n")
# for name, embedding_model in EMBEDDING_CONFIGS.items():
# try:
# print(f"🔍 {name}:")
# print("-" * 50)
# docs = test_retriever_with_embeddings(query, embedding_model, k)
# for i, doc in enumerate(docs, 1):
# source = doc.metadata.get("source", "unknown")
# page = doc.metadata.get("page", "N/A")
# preview = doc.page_content[:300]
# if len(doc.page_content) > 300:
# preview += "..."
# print(f"--- Chunk #{i} ---")
# print(f"Source: {source} | Page: {page}")
# print(preview)
# print()
# print("\n" + "=" * 60 + "\n")
# except Exception as e:
# print(f"❌ Error with {name}: {e}\n")
# if __name__ == "__main__":
# print("Embedding Model Benchmark Tool")
# print("\nType 'compare: <question>' to compare all embeddings")
# print("Type 'exit' to quit\n")
# while True:
# user_input = input("\nEnter your question: ").strip()
# if user_input.lower() == "exit":
# break
# elif user_input.lower().startswith("compare: "):
# query = user_input[9:]
# compare_embeddings(query)
# else:
# print("Please use the format: compare: <question>")
from utils.vector_store import get_vector_store
def test_retriever_with_embeddings(query: str, embedding_model, k: int = 3, vector_store_path="./chroma_db"):
"""Test retriever with a specific embedding model and vector store"""
vector_store = get_vector_store(
persist_directory=vector_store_path, embedding=embedding_model)
retriever = vector_store.as_retriever(search_kwargs={"k": k})
docs = retriever.get_relevant_documents(query)
# Deduplicate based on page_content
seen = set()
unique_docs = []
for doc in docs:
if doc.page_content not in seen:
seen.add(doc.page_content)
unique_docs.append(doc)
print(f"\nUsing vector store: {vector_store_path}")
print(f"Top {len(unique_docs)} unique chunks retrieved for: '{query}'\n")
for i, doc in enumerate(unique_docs, 1):
source = doc.metadata.get("source", "unknown")
page = doc.metadata.get("page", "N/A")
print(f"--- Chunk #{i} ---")
print(f"Source: {source} | Page: {page}")
preview = doc.page_content[:300]
if len(doc.page_content) > 300:
preview += "..."
print(preview)
print()
def compare_retrievers_with_embeddings(query: str, embedding_model, k: int = 3):
"""Compare results from different vector stores using the same embedding model"""
stores = {
"MES Manual": "./vector_stores/mes_db",
"Technical Docs": "./vector_stores/tech_db",
"General Docs": "./vector_stores/general_db"
}
print(f"\n=== Comparing retrievers for: '{query}' ===\n")
for store_name, store_path in stores.items():
try:
print(f"🔍 {store_name}:")
print("-" * 50)
test_retriever_with_embeddings(
query, embedding_model, k=k, vector_store_path=store_path)
print("\n" + "="*60 + "\n")
except Exception as e:
print(f"❌ Could not access {store_name}: {e}\n")
if __name__ == "__main__":
from embedding_config import EMBEDDING_CONFIGS
print("Multi-Vector Store RAG Tester (with Embeddings)")
print("\nAvailable commands:")
print(" - Enter a question to test default store")
print(" - Type 'mes: <question>' for MES manual")
print(" - Type 'tech: <question>' for technical docs")
print(" - Type 'general: <question>' for general docs")
print(" - Type 'compare: <question>' to compare all stores")
print(" - Type 'exit' to quit")
# Choose embedding model at start
print("\nAvailable Embedding Models:")
for i, name in enumerate(EMBEDDING_CONFIGS.keys(), 1):
print(f" {i}. {name}")
choice = int(input("Select embedding model number: ").strip())
embedding_model = list(EMBEDDING_CONFIGS.values())[choice - 1]
while True:
user_input = input("\nEnter your question: ").strip()
if user_input.lower() == "exit":
break
elif user_input.lower().startswith("mes: "):
query = user_input[5:]
test_retriever_with_embeddings(
query, embedding_model, vector_store_path="./vector_stores/mes_db")
elif user_input.lower().startswith("tech: "):
query = user_input[6:]
test_retriever_with_embeddings(
query, embedding_model, vector_store_path="./vector_stores/tech_db")
elif user_input.lower().startswith("general: "):
query = user_input[9:]
test_retriever_with_embeddings(
query, embedding_model, vector_store_path="./vector_stores/general_db")
elif user_input.lower().startswith("compare: "):
query = user_input[9:]
compare_retrievers_with_embeddings(query, embedding_model)
else:
test_retriever_with_embeddings(
user_input, embedding_model) # Default store
|