Files
zeavis-edu/apps/ml-service/README.md
T

294 lines
6.3 KiB
Markdown
Raw Normal View History

2026-05-23 10:51:59 +00:00
# ML Service — Rust Axum ONNX Runtime
Layanan inferensi machine learning berbasis Rust dengan Axum web framework dan ONNX Runtime untuk klasifikasi penyakit daun jagung. Service ini menyediakan endpoint HTTP untuk prediksi real-time dengan performa tinggi dan konsumsi resource minimal.
## Fitur
- **Framework:** Axum (async Rust web framework)
- **Runtime Inferensi:** ONNX Runtime untuk kompatibilitas lintas platform
- **Model:** EfficientNetV2B0 dalam format ONNX
- **Endpoint:** Health check, metadata, dan prediksi gambar
- **Multipart Upload:** Dukungan upload gambar langsung via HTTP POST
## Prasyarat
- Rust 1.70+ dan Cargo
- Model ONNX di `../../Machine_Learning/model/model.onnx` (atau path custom via `MODEL_PATH`)
## Instalasi & Setup
### Instalasi Dependensi
Dependensi Rust sudah terdaftar di `Cargo.toml`. Cargo akan mengunduh dan mengkompilasi otomatis saat pertama kali build.
```bash
cargo build
```
## Menjalankan Service Lokal
2026-05-23 11:05:49 +00:00
Semua perintah di bawah dijalankan dari direktori `apps/ml-service`.
2026-05-23 11:01:41 +00:00
### Opsi 1: Default (Port 8000, Model dari Machine_Learning/)
2026-05-23 10:51:59 +00:00
```bash
2026-05-23 11:05:49 +00:00
cd apps/ml-service
2026-05-23 10:51:59 +00:00
cargo run
```
2026-05-23 11:01:41 +00:00
Service akan mencari model di path default dan mendengarkan di `http://localhost:8000`:
2026-05-23 10:51:59 +00:00
```
../../Machine_Learning/model/model.onnx
```
2026-05-23 11:01:41 +00:00
### Opsi 2: Local Development dengan .env.example (Port 8001)
Untuk development lokal dengan port 8001 (sesuai `.env.example`):
```bash
2026-05-23 11:05:49 +00:00
cd apps/ml-service
2026-05-23 11:01:41 +00:00
source .env.example
cargo run
```
Service akan mendengarkan di `http://localhost:8001` karena `ML_SERVICE_PORT=8001` di `.env.example`.
### Opsi 3: Custom Model Path
2026-05-23 10:51:59 +00:00
Jika model berada di lokasi lain, gunakan environment variable `MODEL_PATH`:
```bash
2026-05-23 11:05:49 +00:00
cd apps/ml-service
2026-05-23 10:51:59 +00:00
MODEL_PATH=/path/to/model.onnx cargo run
```
2026-05-23 11:01:41 +00:00
Atau kombinasikan dengan port custom:
```bash
2026-05-23 11:05:49 +00:00
cd apps/ml-service
2026-05-23 11:01:41 +00:00
ML_SERVICE_PORT=9000 MODEL_PATH=/path/to/model.onnx cargo run
```
2026-05-23 10:51:59 +00:00
## Environment Variables
| Variable | Default | Keterangan |
|---|---|---|
2026-05-23 11:01:41 +00:00
| `ML_SERVICE_HOST` | `0.0.0.0` | Bind address |
| `ML_SERVICE_PORT` | `8000` | Bind port (override untuk local dev dengan `.env.example`) |
2026-05-23 10:51:59 +00:00
| `MODEL_PATH` | `../../Machine_Learning/model/model.onnx` | Path ke file model ONNX |
| `MODEL_INPUT_SIZE` | `224` | Ukuran input gambar (224x224 untuk EfficientNetV2B0) |
| `RUST_LOG` | `info` | Level logging (debug, info, warn, error) |
## Endpoint API
### 1. Health Check
2026-05-23 11:01:41 +00:00
**Default (port 8000):**
2026-05-23 10:51:59 +00:00
```bash
curl http://localhost:8000/health
```
2026-05-23 11:01:41 +00:00
**Local dev dengan .env.example (port 8001):**
```bash
curl http://localhost:8001/health
```
2026-05-23 10:51:59 +00:00
**Response:**
```json
{
"status": "ok",
2026-05-23 11:09:16 +00:00
"model_loaded": true
2026-05-23 10:51:59 +00:00
}
```
### 2. Metadata
2026-05-23 11:01:41 +00:00
**Default (port 8000):**
2026-05-23 10:51:59 +00:00
```bash
curl http://localhost:8000/metadata
```
2026-05-23 11:01:41 +00:00
**Local dev dengan .env.example (port 8001):**
```bash
curl http://localhost:8001/metadata
```
2026-05-23 10:51:59 +00:00
**Response:**
```json
{
2026-05-23 11:09:16 +00:00
"service_name": "zeavis-ml-service",
"service_version": "0.1.0",
2026-05-23 10:51:59 +00:00
"model_path": "../../Machine_Learning/model/model.onnx",
2026-05-23 11:09:16 +00:00
"model_loaded": true,
2026-05-23 10:51:59 +00:00
"input_size": 224,
"labels": [
"Bercak Daun",
2026-05-23 11:09:16 +00:00
"Daun Sehat",
2026-05-23 10:51:59 +00:00
"Karat Daun",
2026-05-23 11:09:16 +00:00
"Hawar Daun"
2026-05-23 10:51:59 +00:00
]
}
```
### 3. Prediksi
Upload gambar daun jagung untuk klasifikasi:
2026-05-23 11:01:41 +00:00
**Default (port 8000):**
2026-05-23 10:51:59 +00:00
```bash
curl -X POST http://localhost:8000/predict \
-F "file=@/path/to/corn-leaf.jpg"
```
2026-05-23 11:01:41 +00:00
**Local dev dengan .env.example (port 8001):**
```bash
curl -X POST http://localhost:8001/predict \
-F "file=@/path/to/corn-leaf.jpg"
```
2026-05-23 10:51:59 +00:00
**Response:**
```json
{
2026-05-23 11:09:16 +00:00
"label": "Daun Sehat",
"confidence": 0.95,
"probabilities": {
"Bercak Daun": 0.02,
"Daun Sehat": 0.95,
"Karat Daun": 0.01,
"Hawar Daun": 0.02
2026-05-23 10:51:59 +00:00
}
}
```
## Verifikasi & Testing
### Build Produksi
```bash
cargo build --release
```
Output binary akan tersedia di `target/release/zeavis-ml-service`.
### Menjalankan Tests
```bash
cargo test
```
Tests mencakup validasi loading model, preprocessing gambar, dan output prediksi.
### Verifikasi Manual
2026-05-23 11:01:41 +00:00
#### Dengan default port 8000:
2026-05-23 10:51:59 +00:00
1. Jalankan service:
```bash
cargo run
```
2. Di terminal lain, test health endpoint:
```bash
curl http://localhost:8000/health
```
3. Test metadata:
```bash
curl http://localhost:8000/metadata
```
4. Test prediksi dengan gambar sample:
```bash
curl -X POST http://localhost:8000/predict \
-F "file=@../../Machine_Learning/dataset/Daun\ Sehat/sample.jpg"
```
2026-05-23 11:01:41 +00:00
#### Dengan local dev port 8001 (.env.example):
1. Jalankan service dengan .env.example:
```bash
source .env.example
cargo run
```
2. Di terminal lain, test health endpoint:
```bash
curl http://localhost:8001/health
```
3. Test metadata:
```bash
curl http://localhost:8001/metadata
```
4. Test prediksi dengan gambar sample:
```bash
curl -X POST http://localhost:8001/predict \
-F "file=@../../Machine_Learning/dataset/Daun\ Sehat/sample.jpg"
```
2026-05-23 10:51:59 +00:00
## Troubleshooting
### Model tidak ditemukan
**Error:** `Failed to load model: No such file or directory`
**Solusi:** Pastikan file model tersedia di path yang benar:
```bash
ls -la ../../Machine_Learning/model/model.onnx
```
Atau set path custom:
```bash
MODEL_PATH=/absolute/path/to/model.onnx cargo run
```
### Port sudah digunakan
**Error:** `Address already in use`
2026-05-23 11:01:41 +00:00
**Solusi:** Service menggunakan port 8000 secara default. Jika port sudah digunakan, ubah dengan environment variable:
```bash
ML_SERVICE_PORT=9000 cargo run
```
Atau jika menggunakan `.env.example` (port 8001), pastikan tidak ada service lain di port tersebut:
```bash
lsof -i :8001
```
2026-05-23 10:51:59 +00:00
### ONNX Runtime tidak kompatibel
**Error:** `ONNX Runtime initialization failed`
**Solusi:** Pastikan ONNX Runtime binary kompatibel dengan sistem operasi. Cargo akan mengunduh binary yang sesuai otomatis. Jika masalah persisten, coba rebuild:
```bash
cargo clean
cargo build
```
## Deployment
### Docker
Service dapat di-deploy via Docker. Dockerfile sudah tersedia di direktori ini.
```bash
docker build -t zeavis-ml-service .
docker run -p 8000:8000 \
-v /path/to/model.onnx:/app/model.onnx \
-e MODEL_PATH=/app/model.onnx \
zeavis-ml-service
```
### Docker Compose
Lihat `docker-compose.yml` di root repository untuk deployment lengkap dengan web, API, dan ML service.
## Dokumentasi Terkait
- [`Machine_Learning/README.md`](../../Machine_Learning/README.md) — Panduan training dan ekspor model ONNX
- [`README.md`](../../README.md) — Dokumentasi proyek utama