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 connect and create dashboards in Grafana?
Creating dashboards in Grafana is a crucial step in a Machine Learning project in order to effectively visualize and make data-driven decisions. First, it is important to establish a connection to Grafana by configuring the necessary parameters and selecting the table where the predictions are stored. This connection provides an interface where you can perform queries to generate the dashboards.
How to select and visualize your data?
- Data Source: Select the table containing your predictions.
- Initial Query: Perform a
SELECTto visualize all the data and corroborate that they are correctly stored. - Relevant Data: Make sure you are storing all the characteristics or fields of interest in your database. This is fundamental to create graphs that contribute to the decision making process.
How to create your first graph in Grafana?
- Add visualization: Choose a visualization suggestion, for example, a horizontal
bar chart. - Configure axes: Configure the X-axis with the customer ID and the Y-axis with the predictions.
- Details and parameters:
- Assign a descriptive name, such as "Predictions labeled distribution based on ID".
- Write a specific description of what the visualization shows.
- Adjust parameters such as
bar widthand bar color. - Select the appropriate variable to classify, in this case,
prediction.
What types of visualizations can you implement with Grafana?
Grafana offers a wide range of chart types that can be implemented according to the needs of the project:
- Time series.
- Bar charts.
- Pie charts (pie charts).
- Charts.
- Histograms.
The different types of charts allow you to experiment with variables and axes, giving you the flexibility to customize your visualizations according to your specific data needs.
How to monitor Machine Learning models with Grafana?
Monitoring is key to ensure the quality of predictions and model characteristics.
- Data monitoring: Implement monitoring to identify possible data leaks or obsolete features.
- Implementation of alerts: Generate alerts from the
alert listin Grafana to detect changes or anomalies in real time. - Automatic retraining: Prepare your Machine Learning pipelines for automatic retraining in case of detecting data drift.
What are the alternatives to Grafana?
While Grafana is powerful, there are other tools you can use for visualization and monitoring:
- Tableau
- Power BI
- Cloud Servers (AWS, Google Cloud, Azure)
Each of these tools offers different capabilities that can complement or replace the use of Grafana, depending on the requirements of your Machine Learning project.
Additional recommendations.
Never forget the importance of:
- Constantly monitor the model metrics to align the results with the business problem.
- Supervise the
data driftvariables to maintain the model's validity. - Keep your tools updated to ensure that each component of your pipeline works optimally.
Contributions 4
Questions 0
para acceder a Grafana, si les pide usuario y password usen: admin y admin
Want to see more contributions, questions and answers from the community?