Spaces:
Runtime error
Runtime error
Update rag_server.py
Browse files- rag_server.py +29 -20
rag_server.py
CHANGED
@@ -1,3 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
from fastapi import FastAPI, Request
|
2 |
from fastapi.responses import JSONResponse, FileResponse, HTMLResponse
|
3 |
from fastapi.staticfiles import StaticFiles
|
@@ -6,8 +17,6 @@ from rag_system import build_rag_chain, ask_question
|
|
6 |
from vector_store import get_embeddings, load_vector_store
|
7 |
from llm_loader import load_llama_model
|
8 |
import uuid
|
9 |
-
import os
|
10 |
-
import shutil
|
11 |
from urllib.parse import urljoin, quote
|
12 |
|
13 |
from fastapi.responses import StreamingResponse
|
@@ -16,17 +25,17 @@ import time
|
|
16 |
|
17 |
app = FastAPI()
|
18 |
|
19 |
-
#
|
20 |
os.makedirs("static/documents", exist_ok=True)
|
21 |
app.mount("/static", StaticFiles(directory="static"), name="static")
|
22 |
|
23 |
-
#
|
24 |
embeddings = get_embeddings(device="cpu")
|
25 |
vectorstore = load_vector_store(embeddings, load_path="vector_db")
|
26 |
llm = load_llama_model()
|
27 |
-
qa_chain = build_rag_chain(llm, vectorstore, language="
|
28 |
|
29 |
-
#
|
30 |
BASE_URL = "http://220.124.155.35:8500"
|
31 |
|
32 |
class Question(BaseModel):
|
@@ -37,7 +46,7 @@ def get_document_url(source_path):
|
|
37 |
return None
|
38 |
filename = os.path.basename(source_path)
|
39 |
dataset_root = os.path.join(os.getcwd(), "dataset")
|
40 |
-
#
|
41 |
found_path = None
|
42 |
for root, dirs, files in os.walk(dataset_root):
|
43 |
if filename in files:
|
@@ -51,13 +60,13 @@ def get_document_url(source_path):
|
|
51 |
return urljoin(BASE_URL, f"/static/documents/{encoded_filename}")
|
52 |
|
53 |
def create_download_link(url, filename):
|
54 |
-
return f'
|
55 |
|
56 |
@app.post("/ask")
|
57 |
def ask(question: Question):
|
58 |
result = ask_question(qa_chain, question.question)
|
59 |
|
60 |
-
#
|
61 |
sources = []
|
62 |
for doc in result["source_documents"]:
|
63 |
source_path = doc.metadata.get('source', 'N/A')
|
@@ -100,7 +109,7 @@ async def openai_compatible_chat(request: Request):
|
|
100 |
result = ask_question(qa_chain, user_input)
|
101 |
answer = result['result']
|
102 |
|
103 |
-
#
|
104 |
sources = []
|
105 |
for doc in result["source_documents"]:
|
106 |
source_path = doc.metadata.get('source', 'N/A')
|
@@ -116,13 +125,13 @@ async def openai_compatible_chat(request: Request):
|
|
116 |
}
|
117 |
sources.append(source_info)
|
118 |
|
119 |
-
#
|
120 |
-
sources_md = "\
|
121 |
seen = set()
|
122 |
for source in sources:
|
123 |
key = (source['filename'], source['document_url'])
|
124 |
if source['document_url'] and source['filename'] and key not in seen:
|
125 |
-
sources_md += f"
|
126 |
seen.add(key)
|
127 |
|
128 |
final_answer = answer.split("A:")[-1].strip() if "A:" in answer else answer.strip()
|
@@ -143,9 +152,9 @@ async def openai_compatible_chat(request: Request):
|
|
143 |
"model": "rag",
|
144 |
})
|
145 |
|
146 |
-
#
|
147 |
def event_stream():
|
148 |
-
#
|
149 |
answer_main = answer.split("A:")[-1].strip() if "A:" in answer else answer.strip()
|
150 |
for char in answer_main:
|
151 |
chunk = {
|
@@ -161,15 +170,15 @@ async def openai_compatible_chat(request: Request):
|
|
161 |
}
|
162 |
yield f"data: {json.dumps(chunk)}\n\n"
|
163 |
time.sleep(0.005)
|
164 |
-
#
|
165 |
-
sources_md = "\
|
166 |
seen = set()
|
167 |
for source in sources:
|
168 |
key = (source['filename'], source['document_url'])
|
169 |
if source['document_url'] and source['filename'] and key not in seen:
|
170 |
-
sources_md += f"
|
171 |
seen.add(key)
|
172 |
-
if sources_md.strip() != "
|
173 |
chunk = {
|
174 |
"id": f"chatcmpl-{uuid.uuid4()}",
|
175 |
"object": "chat.completion.chunk",
|
@@ -194,4 +203,4 @@ async def openai_compatible_chat(request: Request):
|
|
194 |
yield f"data: {json.dumps(done)}\n\n"
|
195 |
return
|
196 |
|
197 |
-
return StreamingResponse(event_stream(), media_type="text/event-stream")
|
|
|
1 |
+
import os
|
2 |
+
import re
|
3 |
+
import glob
|
4 |
+
import time
|
5 |
+
from collections import defaultdict
|
6 |
+
|
7 |
+
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
8 |
+
from langchain_core.documents import Document
|
9 |
+
from langchain_community.embeddings import HuggingFaceEmbeddings
|
10 |
+
from langchain_community.vectorstores import FAISS
|
11 |
+
|
12 |
from fastapi import FastAPI, Request
|
13 |
from fastapi.responses import JSONResponse, FileResponse, HTMLResponse
|
14 |
from fastapi.staticfiles import StaticFiles
|
|
|
17 |
from vector_store import get_embeddings, load_vector_store
|
18 |
from llm_loader import load_llama_model
|
19 |
import uuid
|
|
|
|
|
20 |
from urllib.parse import urljoin, quote
|
21 |
|
22 |
from fastapi.responses import StreamingResponse
|
|
|
25 |
|
26 |
app = FastAPI()
|
27 |
|
28 |
+
# Configuration for serving static files
|
29 |
os.makedirs("static/documents", exist_ok=True)
|
30 |
app.mount("/static", StaticFiles(directory="static"), name="static")
|
31 |
|
32 |
+
# Prepare global objects
|
33 |
embeddings = get_embeddings(device="cpu")
|
34 |
vectorstore = load_vector_store(embeddings, load_path="vector_db")
|
35 |
llm = load_llama_model()
|
36 |
+
qa_chain = build_rag_chain(llm, vectorstore, language="en", k=7)
|
37 |
|
38 |
+
# Server URL configuration (adjust to match your actual environment)
|
39 |
BASE_URL = "http://220.124.155.35:8500"
|
40 |
|
41 |
class Question(BaseModel):
|
|
|
46 |
return None
|
47 |
filename = os.path.basename(source_path)
|
48 |
dataset_root = os.path.join(os.getcwd(), "dataset")
|
49 |
+
# Find file matching filename in the entire dataset subdirectory
|
50 |
found_path = None
|
51 |
for root, dirs, files in os.walk(dataset_root):
|
52 |
if filename in files:
|
|
|
60 |
return urljoin(BASE_URL, f"/static/documents/{encoded_filename}")
|
61 |
|
62 |
def create_download_link(url, filename):
|
63 |
+
return f'Source: [{filename}]({url})'
|
64 |
|
65 |
@app.post("/ask")
|
66 |
def ask(question: Question):
|
67 |
result = ask_question(qa_chain, question.question)
|
68 |
|
69 |
+
# Process source document information
|
70 |
sources = []
|
71 |
for doc in result["source_documents"]:
|
72 |
source_path = doc.metadata.get('source', 'N/A')
|
|
|
109 |
result = ask_question(qa_chain, user_input)
|
110 |
answer = result['result']
|
111 |
|
112 |
+
# Process source document information
|
113 |
sources = []
|
114 |
for doc in result["source_documents"]:
|
115 |
source_path = doc.metadata.get('source', 'N/A')
|
|
|
125 |
}
|
126 |
sources.append(source_info)
|
127 |
|
128 |
+
# Output source information one line at a time
|
129 |
+
sources_md = "\nReferences Documents:\n"
|
130 |
seen = set()
|
131 |
for source in sources:
|
132 |
key = (source['filename'], source['document_url'])
|
133 |
if source['document_url'] and source['filename'] and key not in seen:
|
134 |
+
sources_md += f"Source: [{source['filename']}]({source['document_url']})\n"
|
135 |
seen.add(key)
|
136 |
|
137 |
final_answer = answer.split("A:")[-1].strip() if "A:" in answer else answer.strip()
|
|
|
152 |
"model": "rag",
|
153 |
})
|
154 |
|
155 |
+
# Generator for streaming response
|
156 |
def event_stream():
|
157 |
+
# Stream only the answer body first
|
158 |
answer_main = answer.split("A:")[-1].strip() if "A:" in answer else answer.strip()
|
159 |
for char in answer_main:
|
160 |
chunk = {
|
|
|
170 |
}
|
171 |
yield f"data: {json.dumps(chunk)}\n\n"
|
172 |
time.sleep(0.005)
|
173 |
+
# Send reference documents (download links) all at once at the end
|
174 |
+
sources_md = "\nReferences Documents:\n"
|
175 |
seen = set()
|
176 |
for source in sources:
|
177 |
key = (source['filename'], source['document_url'])
|
178 |
if source['document_url'] and source['filename'] and key not in seen:
|
179 |
+
sources_md += f"Source: [{source['filename']}]({source['document_url']})\n"
|
180 |
seen.add(key)
|
181 |
+
if sources_md.strip() != "References Documents:":
|
182 |
chunk = {
|
183 |
"id": f"chatcmpl-{uuid.uuid4()}",
|
184 |
"object": "chat.completion.chunk",
|
|
|
203 |
yield f"data: {json.dumps(done)}\n\n"
|
204 |
return
|
205 |
|
206 |
+
return StreamingResponse(event_stream(), media_type="text/event-stream")
|