Spaces:
Sleeping
Sleeping
Update util.py
Browse files
util.py
CHANGED
@@ -1,51 +1,51 @@
|
|
1 |
-
import os
|
2 |
-
import json
|
3 |
-
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
4 |
-
from langchain_community.vectorstores.faiss import FAISS
|
5 |
-
from langchain.chains.question_answering import load_qa_chain
|
6 |
-
from langchain.prompts import PromptTemplate
|
7 |
-
from langchain_google_genai import GoogleGenerativeAIEmbeddings
|
8 |
-
from langchain_google_genai import ChatGoogleGenerativeAI
|
9 |
-
import google.generativeai as genai
|
10 |
-
from dotenv import load_dotenv
|
11 |
-
|
12 |
-
|
13 |
-
|
14 |
-
|
15 |
-
|
16 |
-
|
17 |
-
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
|
22 |
-
|
23 |
-
|
24 |
-
|
25 |
-
|
26 |
-
|
27 |
-
|
28 |
-
rating
|
29 |
-
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
|
37 |
-
|
38 |
-
|
39 |
-
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
|
44 |
-
|
45 |
-
|
46 |
-
text = extract_text(json_path)
|
47 |
-
chunks = split_text_into_chunks(text)
|
48 |
-
create_vector_store(chunks)
|
49 |
-
|
50 |
-
json_path = 'reviews.json'
|
51 |
main(json_path)
|
|
|
1 |
+
import os
|
2 |
+
import json
|
3 |
+
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
4 |
+
from langchain_community.vectorstores.faiss import FAISS
|
5 |
+
from langchain.chains.question_answering import load_qa_chain
|
6 |
+
from langchain.prompts import PromptTemplate
|
7 |
+
from langchain_google_genai import GoogleGenerativeAIEmbeddings
|
8 |
+
from langchain_google_genai import ChatGoogleGenerativeAI
|
9 |
+
import google.generativeai as genai
|
10 |
+
from dotenv import load_dotenv
|
11 |
+
|
12 |
+
def extract_text(json_path):
|
13 |
+
with open(json_path, 'r') as file:
|
14 |
+
data = json.load(file)
|
15 |
+
|
16 |
+
text = ""
|
17 |
+
for professor in data['professors']:
|
18 |
+
professor_id = professor.get('professor_id')
|
19 |
+
name = professor.get('name')
|
20 |
+
course = professor.get('course')
|
21 |
+
reviews = professor.get('reviews', [])
|
22 |
+
|
23 |
+
text += f'\nProfessor ID: {professor_id}, Professor Name: {name}, Course: {course}\n '
|
24 |
+
if reviews:
|
25 |
+
for review in reviews:
|
26 |
+
rating = review.get('rating')
|
27 |
+
review_text = review.get('review_text')
|
28 |
+
text += f"Rating: {rating}, Review: {review_text}\n"
|
29 |
+
else:
|
30 |
+
print("No reviews available.")
|
31 |
+
return text
|
32 |
+
|
33 |
+
def split_text_into_chunks(text):
|
34 |
+
splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
|
35 |
+
text_chunks = splitter.split_text(text)
|
36 |
+
return text_chunks
|
37 |
+
|
38 |
+
def create_vector_store(chunks):
|
39 |
+
embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
|
40 |
+
vector_store = FAISS.from_texts(chunks, embedding=embeddings)
|
41 |
+
vector_store.save_local("reviews_index")
|
42 |
+
|
43 |
+
def main(json_path):
|
44 |
+
load_dotenv()
|
45 |
+
genai.configure(api_key=os.getenv("GOOGLE_API_KEY"))
|
46 |
+
text = extract_text(json_path)
|
47 |
+
chunks = split_text_into_chunks(text)
|
48 |
+
create_vector_store(chunks)
|
49 |
+
|
50 |
+
json_path = 'reviews.json'
|
51 |
main(json_path)
|