A robust middleware for hybrid semantic caching, text normalization, and vector search optimization.
Project description
======================================================================== SOURCE CODE SKRIPSI: HYBRID SEMANTIC CACHE MIDDLEWARE Oleh: Sandy Wicaksono
A. DESKRIPSI PROYEK Proyek ini adalah implementasi arsitektur Hybrid Semantic Cache Middleware menggunakan model embedding all-MiniLM-L6-v2 dan database vektor FAISS. Middleware ini bertujuan untuk mencegat kueri bahasa gaul/typo dari pengguna (melalui modul Normalizer) dan membalasnya secara lokal untuk memotong latensi (waktu respons) dari LLM Google Gemini 3.1 Flash Lite di Cloud.
B. PRASYARAT (PREREQUISITES)
- Python versi 3.10 atau 3.11 telah terinstal.
- Memiliki API Key dari Google Gemini (Google AI Studio).
C. CARA INSTALASI
- Buka Terminal / Command Prompt di dalam folder proyek ini.
- (Opsional) Buat virtual environment:
python -m venv env_skripsi
Lalu aktifkan:
- Windows: .\env_skripsi\Scripts\activate
- Mac/Linux: source env_skripsi/bin/activate
- Instal semua library yang dibutuhkan dengan perintah: pip install -r requirements.txt
D. PENGATURAN API KEY GEMINI
Buka file src/gemini_api.py dan masukkan API Key Gemini Anda pada variabel:
API_KEY = "MASUKKAN_API_KEY_ANDA_DI_SINI"
E. CARA MENJALANKAN SERVER MIDDLEWARE (FASTAPI)
- Buka terminal di direktori utama proyek.
- Jalankan perintah berikut: uvicorn src.main:app --reload
- Server lokal akan berjalan di http://127.0.0.1:8000
- Anda dapat mencoba endpoint dengan membuka URL Swagger UI di: http://127.0.0.1:8000/docs
F. CARA MENJALANKAN PENGUJIAN (BAB 4 SKRIPSI) Untuk menjalankan simulasi pengujian Skenario A, B, dan C (Hit Rate & Latency):
- Buka terminal baru.
- Jalankan script pengujian: python evaluasi/test_scenarios.py
- Untuk membuat grafik visualisasi hasil pengujian, jalankan: python evaluasi/generate_grafik.py
- File gambar grafik (.png) akan otomatis tersimpan di dalam folder
output/.
======================================================================== Terima kasih. Jika ada kendala dalam menjalankan program, silakan menghubungi peneliti (Sandy).
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file hybrid_semantic_cache-0.1.0.tar.gz.
File metadata
- Download URL: hybrid_semantic_cache-0.1.0.tar.gz
- Upload date:
- Size: 6.6 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
1af71cc3b615e6b1f182bc92228f58180c90ee8cac78ee8386e36292b39b380a
|
|
| MD5 |
36e1120c64229105a13be02a296c8b3d
|
|
| BLAKE2b-256 |
ac6200caa64feb0fe2899809c9af6aac6347a39fc91e49025d5995ff771b2d15
|
File details
Details for the file hybrid_semantic_cache-0.1.0-py3-none-any.whl.
File metadata
- Download URL: hybrid_semantic_cache-0.1.0-py3-none-any.whl
- Upload date:
- Size: 7.7 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
b624a9f037a6a3b94800167e31c4526874d159d068d09311a3ebd02ed00e510e
|
|
| MD5 |
0fa9f89d9ad9720940074f925fd88d75
|
|
| BLAKE2b-256 |
a10fae9f1211e1d54e28836f41a8bd6568ae108be40ec52f1146e95d37df4455
|