Fundamentos de MLOps y tracking de modelos
¿Qué es MLOps y para qué sirve?
Tracking de modelos en localhost con MLflow
Tracking de modelos en localhost: directorio personalizado
Etapas del ciclo de MLOps
Componentes de MLOps
Tracking de modelos con MLflow y SQLite
Tracking de modelos con MLflow en la nube
Tracking del ciclo de vida de modelos de machine learning
Tracking de experimentos con MLflow: preprocesamiento de datos
Tracking de experimentos con MLflow: definición de funciones
Tracking de experimentos con MLflow: tracking de métricas e hiperparámetros
Tracking de experimentos con MLflow: reporte de clasificación
Entrenamiento de modelos baseline y análisis en UI de MLflow
MLflow Model Registry: registro y uso de modelos
Registro de modelos con mlflow.client
Testing de modelo desde MLflow con datos de prueba
¿Para qué sirve el tracking de modelos en MLOps?
Orquestación de pipelines de machine learning
Tasks con Prefect
Flows con Prefect
Flow de modelo de clasificación de tickets: procesamiento de datos y features
Flow de modelo de clasificación de tickets: integración de las tasks
Flow de modelo de clasificación de tickets: ejecución de tasks
¿Cómo se integra la orquestación en MLOps?
Despliegue de modelo de machine learning
Despligue con Docker y FastAPI: configuración y requerimientos
Despligue con Docker y FastAPI: definición de clases y entry point
Despligue con Docker y FastAPI: procesamiento de predicciones en main app
Despligue con Docker y FastAPI: configuración de la base de datos
Despliegue y pruebas de modelo de machine learning en localhost
Despliegue y pruebas de modelo de machine learning en la nube
¿Qué hacer con el modelo desplegado?
Monitoreo de modelo de machine learning en producción
¿Cómo monitorear modelos de machine learning en producción?
Entrenamiento de modelo baseline
Preparar datos para crear reporte con Evidently
Análisis de la calidad de los datos con Evidently
Creación de reportes con Grafana
¿Cómo mejorar tus procesos de MLOps?
You don't have access to this class
Keep learning! Join and start boosting your career
If you already have an account,
What to do after deploying a machine learning model in production?
Once you have deployed a machine learning model in production, the work does not end there. It is essential to ensure a series of actions and measures to maintain the efficiency and continuity of the service. Here we break down the most essential points to keep in mind.
How to ensure adequate monitoring?
Constant monitoring of the model is vital for its optimal operation. This not only implies keeping an eye on the model's behavior, but also taking preventive and corrective actions to ensure its stability.
- Implement an alert system in case the model's performance declines.
- Periodically review system logs to identify patterns or anomalies.
- Continually evaluates model results through predefined metrics to detect deviations.
What role does the infrastructure play in production?
The infrastructure on which the model is deployed is crucial, especially when there is an increase in the number of customer requests. If you are not prepared to support these requirements, the service could fail. To prevent this:
- Auto-scaling: ensures that the infrastructure can automatically scale with the increase in traffic. Use cloud services that offer this capability.
- Keep CI/CD (Continuous Integration and Continuous Delivery) pipelines up to date to incorporate changes without interrupting service.
- Optimize the use of cloud resources, which will allow you to extend them when necessary and avoid service interruptions.
What other recommendations are key?
To further improve the deployment of models in a production environment, consider the following recommendations:
- Infrastructure requirements: Ensure that the infrastructure meets the necessary requirements based on the serving, i.e., how the model is delivered to the end user.
- Updating workflows: Keep workflows up to date to accommodate necessary changes or enhancements. This includes both software and hardware.
- Reference material: Take advantage of additional materials, such as downloadable documentation on model deployment, to review and consolidate your knowledge.
Actively participating in the monitoring and maintenance of a deployed model will help you minimize risk, optimize resources and ensure continued business success. With these strategies well implemented, you will be better prepared for challenges that may arise along the way. Always remember that learning and adaptation are key in this exciting world of machine learning in production.
Contributions 0
Questions 1
Want to see more contributions, questions and answers from the community?