Google Cloud Professional Machine Learning Engineer Practice Test
Examen: Google Cloud Professional Machine Learning Engineer Descripción: Examen de práctica para la certificación Google Cloud Professional Machine Learning Engineer. Cubre el diseño de soluciones de ML, Vertex AI, pipelines de ML, implementación de modelos, monitoreo y MLOps en Google Cloud.
Examen de certificación
Professional
Nivel
Oportunidades profesionales y salario
stable mercado
Plan de Estudio
Cada dominio está ponderado para coincidir con el examen de certificación real, por lo que una simulación de práctica completa predice tu resultado.
01Architecting Low-Code ML Solutions
02Automating and Orchestrating ML PipelinesVertex AI Pipelines (KFP v2): components, artifacts, parameters, DAG parallelism, caching, retry, TFX pipeline components and TFMD metadata, Cloud Composer (Airflow) for multi-service orchestration, MLOps maturity levels (Level 0/1/2): CI, CD, Continuous Training, Cloud Build for ML CI pipelines, Eventarc and Cloud Scheduler for pipeline triggers
20-24%
03Collaborating within and Across TeamsVertex AI Feature Store (online and batch serving, point-in-time retrieval), Vertex AI Model Registry (versioning, rollback, Model Cards), Vertex AI Experiments (run tracking, comparison), Vertex AI ML Metadata and artifact lineage, IAM and data governance (BigQuery authorized views, Dataplex, VPC Service Controls), Cloud Workstations (2023) and Vertex AI Workbench
12-16%
04Monitoring Model Behavior
05Scaling Prototypes into ML ModelsContainerizing training code (custom containers, pre-built DLCs), Distributed training (MirroredStrategy, MultiWorkerMirroredStrategy, TPUStrategy, Horovod), Cloud TPU vs GPU selection, Vertex AI Hyperparameter Tuning (Bayesian, Random, Grid) and Vizier, Feature engineering (normalization, standardization, encoding, embeddings, interaction features, target encoding), Bias-variance tradeoff, regularization (L1/L2/ElasticNet), cross-validation
16-20%
06Serving and Scaling ModelsVertex AI Endpoints (dedicated vs shared, autoscaling, min/max replicas), Online prediction, batch prediction, traffic split, canary deployment, Cloud Functions and Cloud Run for model serving, Vertex AI Batch Prediction for large-scale scoring, Private Service Connect for VPC-only endpoint access, Preprocessing in SavedModel serving signature
18-22%
Preguntas Frecuentes
Reseñas y Calificaciones
Aún sin reseñas
¡Sé el primero en reseñar este examen y ayuda a otros estudiantes!
Comparte Tu Experiencia
Inicia sesión para dejar una reseña
Comparte tu experiencia y ayuda a otros estudiantes a tomar decisiones informadas.
¿No ves tu examen?
Crea un examen de práctica con tu propio material de estudio.