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,
How to improve your MLOps processes?
Reaching the end of a course is always an impressive achievement, especially when it comes to such complex and rapidly developing topics as MLOps. You're in the right place if you're looking for guidance on how to take your skills to the next level. The goal is not only to maintain efficiency, but to constantly increase the scalability and functionality of your machine learning services. Here are some tips that will help you achieve this.
Why is infrastructure as code important?
Infrastructure as code is critical to scaling your services and ensuring that they remain operational in the face of increased requests. Tools like Terraform can be a good starting point. The most important thing to understand is that it is not enough to make a service available; you must always be prepared for unexpected changes in demand.
Other aspects to consider are:
-
Scalability and resource management: Make sure that your infrastructure can automatically adapt to variations in traffic. This involves techniques that allow continuous availability of your interconnected services.
-
Specialized monitoring: Develop skills in specialized monitoring techniques that vary according to the type of data you handle. The tools for structured data will not always be the same for unstructured data.
What have you learned about MLOps?
Throughout this course, you've gained essential knowledge about MLOps, from why its application is crucial to how to maintain the lifecycle of machine learning models. Here are some key points:
-
Model Tracking: Whether locally, using a back-end, or through cloud services like AWS, you know how to track effectively.
-
Task Orchestration: You are able to integrate logical flows by creating MLOps pipelines for effective task orchestration.
-
Model deployment: You have learned how to deploy batch models and the infrastructure needed to make them available as services.
-
Monitoring models in production: You know how to monitor models in production and know the tips and recommendations to ensure their optimal performance.
How can I apply this knowledge in practice?
The theory you have learned is just the beginning. It is crucial that you now apply these concepts in real machine learning projects. Here are some recommendations:
-
Create and share: Develop and share the project you started during this course. This will not only strengthen your skills but also provide you with valuable feedback.
-
Integrate MLOps into new projects: Look for opportunities to integrate MLOps into different stages of your projects, ensuring efficiency and effectiveness.
Remember that learning is a continuous journey. Technology is always advancing and, with it, our skills must be updated. Keep exploring, experimenting and never stop learning. I am sure that with dedication and curiosity, you will achieve great things in the world of machine learning. Go ahead!
Contributions 4
Questions 0
He llegado hasta acá con un sabor de boca agridulce.
Ya que era un curso que esperaba desde hace mucho y peleaba en su momento.
-
Lo dulce: Antes que nada, no tendría ninguna queja para la profesora. Su forma de enseñar y su conocimiento sobre el área: 10/10.
-
Lo agrio: La parte agria es el esfuerzo por parte del team de platzi al diseñar y editar esté curso. Hay videos con delay, otros videos que están saltados, otro caso por ejemplo; es que mencionan que dejarán algo en los recursos y no hay nada. Entre otras cosas que me decepcionaron.
Por mi parte ya metí la queja con el @teamplatzi, pero no he tenido éxito de que hayan hecho algo al respecto.
Genial Este curso!!! Muy organizado con la documentación. Se ve la preparación del curso. un curso de gran valor para mi perfil profesional!!! Gracias
Want to see more contributions, questions and answers from the community?