Detect drift, track performance, monitor data quality and receive alerts using one simple Python library.
Drift Methods
Metrics
Configuration
Open Source
Detect changes in your data using KS Test, PSI and Mean Shift.
Track Accuracy, F1 Score, MAE, RMSE and model degradation.
Find null values, outliers and schema issues before deployment.
Receive notifications through webhooks and custom callbacks.
Start monitoring your models immediately.
import mlwatch
monitor = mlwatch.Monitor(
name="fraud_detector"
)
result = monitor.log(
reference=train_data,
current=new_data
)
print(result.to_dict())| Feature | MLWatch | Evidently | WhyLogs |
|---|---|---|---|
| Lightweight | ✅ | ❌ | ❌ |
| Simple API | ✅ | ❌ | ❌ |
| JSON Output | ✅ | ❌ | ❌ |
| SQLite History | ✅ | ❌ | ❌ |