forked from PromtEngineer/localGPT
-
Notifications
You must be signed in to change notification settings - Fork 0
/
run_localGPT_API.py
162 lines (129 loc) · 4.75 KB
/
run_localGPT_API.py
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
import logging
import os
import shutil
import subprocess
import argparse
# import torch
from flask import Flask, jsonify, request
from langchain.chains import RetrievalQA
from utils import get_embeddings, get_llm, get_db
from prompt_template_utils import get_prompt_template
# from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
from werkzeug.utils import secure_filename
# API queue addition
from threading import Lock
request_lock = Lock()
SHOW_SOURCES = True
logging.info(f"Display Source Documents set to: {SHOW_SOURCES}")
EMBEDDINGS = get_embeddings()
# load the vectorstore
DB = get_db()
RETRIEVER = DB.as_retriever()
LLM = get_llm()
prompt, memory = get_prompt_template(promptTemplate_type="llama", history=False)
QA = RetrievalQA.from_chain_type(
llm=LLM,
chain_type="stuff",
retriever=RETRIEVER,
return_source_documents=SHOW_SOURCES,
chain_type_kwargs={
"prompt": prompt,
},
)
app = Flask(__name__)
@app.route("/api/delete_source", methods=["GET"])
def delete_source_route():
folder_name = "SOURCE_DOCUMENTS"
if os.path.exists(folder_name):
shutil.rmtree(folder_name)
os.makedirs(folder_name)
return jsonify({"message": f"Folder '{folder_name}' successfully deleted and recreated."})
@app.route("/api/save_document", methods=["GET", "POST"])
def save_document_route():
if "document" not in request.files:
return "No document part", 400
file = request.files["document"]
if file.filename == "":
return "No selected file", 400
if file:
filename = secure_filename(file.filename)
folder_path = "SOURCE_DOCUMENTS"
if not os.path.exists(folder_path):
os.makedirs(folder_path)
file_path = os.path.join(folder_path, filename)
file.save(file_path)
return "File saved successfully", 200
@app.route("/api/run_ingest", methods=["GET"])
def run_ingest_route():
global DB
global RETRIEVER
global QA
try:
if os.path.exists(PERSIST_DIRECTORY):
try:
shutil.rmtree(PERSIST_DIRECTORY)
except OSError as e:
print(f"Error: {e.filename} - {e.strerror}.")
else:
print("The directory does not exist")
run_langest_commands = ["python", "ingest.py"]
if DEVICE_TYPE == "cpu":
run_langest_commands.append("--device_type")
run_langest_commands.append(DEVICE_TYPE)
result = subprocess.run(run_langest_commands, capture_output=True)
if result.returncode != 0:
return "Script execution failed: {}".format(result.stderr.decode("utf-8")), 500
# load the vectorstore
DB = get_db()
RETRIEVER = DB.as_retriever()
prompt, memory = get_prompt_template(promptTemplate_type="llama", history=False)
QA = RetrievalQA.from_chain_type(
llm=LLM,
chain_type="stuff",
retriever=RETRIEVER,
return_source_documents=SHOW_SOURCES,
chain_type_kwargs={
"prompt": prompt,
},
)
return "Script executed successfully: {}".format(result.stdout.decode("utf-8")), 200
except Exception as e:
return f"Error occurred: {str(e)}", 500
@app.route("/api/prompt_route", methods=["GET", "POST"])
def prompt_route():
global QA
global request_lock # Make sure to use the global lock instance
user_prompt = request.form.get("user_prompt")
if user_prompt:
# Acquire the lock before processing the prompt
with request_lock:
# print(f'User Prompt: {user_prompt}')
# Get the answer from the chain
res = QA.invoke(user_prompt)
answer, docs = res["result"], res["source_documents"]
prompt_response_dict = {
"Prompt": user_prompt,
"Answer": answer,
}
prompt_response_dict["Sources"] = []
for document in docs:
prompt_response_dict["Sources"].append(
(os.path.basename(str(document.metadata["source"])), str(document.page_content))
)
return jsonify(prompt_response_dict), 200
else:
return "No user prompt received", 400
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument("--port", type=int, default=5110, help="Port to run the API on. Defaults to 5110.")
parser.add_argument(
"--host",
type=str,
default="0.0.0.0",
help="Host to run the UI on. Defaults to 0.0.0.0.",
)
args = parser.parse_args()
logging.basicConfig(
format="%(asctime)s - %(levelname)s - %(filename)s:%(lineno)s - %(message)s", level=logging.INFO
)
app.run(debug=False, host=args.host, port=args.port)