#77 feat: title과 review를 embedding해서 pinecone index에 저장

backend/src/main/java/com/project/capstone/book/domain/Book.java

@@ -27,7 +27,6 @@ public class Book {
     private Long id;
     private String isbn;
     private String title;
-    private String description;
     private String author;
     private String publisher;
     @Column(name = "publish_date")
@@ -52,7 +51,7 @@ public class Book {
     private List<MyBook> membersAddThisBook = new ArrayList<>();
 
     public Book(AddBookRequest request) {
-        this(null, request.isbn(), request.title(), request.description(), request.author(), request.publisher(), request.publishDate(),
+        this(null, request.isbn(), request.title(), request.author(), request.publisher(), request.publishDate(),
                 request.imageUrl(), new ArrayList<>(), new ArrayList<>(), new ArrayList<>(), new ArrayList<>());
     }
 }

fastapi/.gitignore

@@ -0,0 +1,3 @@
+.idea
+.env
+__pycache__

fastapi/main.py

@@ -0,0 +1,58 @@
+import os
+from fastapi import FastAPI
+from pydantic import BaseModel
+from openai import OpenAI
+from pinecone import Pinecone
+from dotenv import load_dotenv
+
+app = FastAPI()
+load_dotenv()
+class EmbeddingRequest(BaseModel):
+    isbn: str
+    title: str
+    description: str
+
+@app.post("/embed")
+async def get_embedding(embedding_request: EmbeddingRequest):
+    isbn = embedding_request.isbn
+    title = embedding_request.title
+    review = embedding_request.description
+
+    # using openai embedding model
+    client = OpenAI(api_key=os.environ.get("OPENAI_API_KEY"))
+    response = client.embeddings.create(
+        input=title,
+        model="text-embedding-3-small"
+    )
+    title_embedding = response.data[0].embedding
+
+    response = client.embeddings.create(
+        input=review,
+        model="text-embedding-3-small"
+    )
+    review_embedding = response.data[0].embedding
+
+    # upsert to pinecone index
+    pc = Pinecone(api_key=os.environ.get("PINECONE_API_KEY"))
+    index = pc.Index("review-and-title-embedding")
+    index.upsert(
+        vectors=[
+            {
+                "id": isbn,
+                "values": title_embedding,
+                "metadata": {"type": "title"}
+            }
+        ],
+        namespace="title-embeddings"
+    )
+    index.upsert(
+        vectors=[
+            {
+                "id": isbn,
+                "values": review_embedding,
+                "metadata": {"type": "review"}
+            }
+        ],
+        namespace="review-embeddings"
+    )
+    return "ok"