feat: client and api

Author: mathhulk

Diff for: .gitignore

@@ -1,3 +1,2 @@
 .DS_Store
-__pycache__
-.env
+.vscode

Diff for: .vscode/settings.json

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

Diff for: packages/api/.gitignore

@@ -15,6 +15,8 @@ python-dotenv = "^1.0.1"
 waitress = "^3.0.0"
 flask-cors = "^5.0.0"
 faiss-cpu = "^1.9.0"
+gdown = "^5.2.0"
+pandas = "^2.2.3"
 
 [tool.poetry.scripts]
 dev = "semantic_search:dev"

Diff for: README.md renamed to packages/api/README.md

@@ -1,23 +1,24 @@
 import os
 import faiss
 import json
-import numpy as np
 from sentence_transformers import SentenceTransformer
 
 os.environ["KMP_DUPLICATE_LIB_OK"] = "TRUE"
 
-index = faiss.read_index('semantic_search/models/atharv-naidu-index.index')
+index = faiss.read_index("semantic_search/models/atharv-naidu-index.index")
 
-with open('semantic_search/models/atharv-naidu-metadata.json', 'r') as f:
+with open("semantic_search/models/atharv-naidu-metadata.json", "r") as f:
     courses = json.load(f)
 
 # Load the SentenceTransformer model
-model = SentenceTransformer('all-MiniLM-L6-v2')
+model = SentenceTransformer("all-MiniLM-L6-v2")
+
 
 # Function to get embeddings for query
 def get_embeddings(text):
     return model.encode(text, convert_to_tensor=True).cpu().numpy().reshape(1, -1)
 
+
 # Function to search courses and return them in the specified format
 def getCourses(query: str, topK: int):
     query_embedding = get_embeddings(query)
@@ -29,7 +30,7 @@ def getCourses(query: str, topK: int):
     results = []
     for idx, dist in zip(indices[0], distances[0]):
         course = courses[idx]
-        
+
         course_result = {
             "subject": course["subject"],
             "number": course["number"],
@@ -38,10 +39,3 @@ def getCourses(query: str, topK: int):
         results.append(course_result)
 
     return results
-
-# Example query
-user_query = "data structures"
-matched_courses = getCourses(user_query, topK=5)
-
-for course in matched_courses:
-    print(course)