Professional Machine Learning Engineer Certification

Professional Machine Learning Engineer Exam Prep Workshop

DreamsPlus presents a Professional Machine Learning Engineer Boot Camp in Singapore and online, tailored to provide practical, hands-on training for aspiring machine learning professionals. This program prepares you for the prestigious Google Professional Machine Learning Engineer certification. With guidance from experienced trainers, you’ll master cutting-edge machine learning concepts, tools, and best practices, empowering you to excel in your certification exam and thrive in Singapore’s dynamic tech industry.

Learning Pathway:

Section 1: Architecting low-code ML solutions 

1.1 Developing ML models by using BigQuery ML. 

• Selecting the appropriate BigQuery ML model (e.g., linear and binary classification, regression, time-series, matrix factorization, boosted trees, autoencoders) based on the business problem.
• Performing feature engineering or feature selection using BigQuery ML.
• Generating predictions using BigQuery ML.

1.2 Building AI solutions by using ML APIs. 

• Developing applications using ML APIs (e.g., Cloud Vision API, Natural Language API, Cloud Speech API, Translation).
• Developing applications using industry-specific APIs (e.g., Document AI API, Retail API).

1.3 Training models by using AutoML. 

• Preparing data for AutoML (e.g., feature selection, data labeling, Tabular Workflows on AutoML).
• Utilizing available data (e.g., tabular, text, speech, images, videos) to train custom models.
• Using AutoML for tabular data.
• Creating forecasting models with AutoML.
• Configuring and troubleshooting trained models.

Section 2: Collaborating within and across teams to manage data and models 

2.1 Exploring and preprocessing organization-wide data (e.g., Cloud Storage, BigQuery, Spanner, Cloud SQL, Apache Spark, Apache Hadoop). Considerations include:

• Organizing various types of data (e.g., tabular, text, speech, images, videos) for optimal training.
• Managing datasets within Vertex AI.
• Performing data preprocessing (e.g., Dataflow, TensorFlow Extended [TFX], BigQuery).
• Creating and managing features in Vertex AI Feature Store.
• Addressing privacy considerations related to data usage and collection (e.g., managing sensitive data such as personally identifiable information [PII] and protected health information [PHI]).

2.2 Model prototyping using Jupyter notebooks. Considerations include:

• Selecting the appropriate Jupyter backend on Google Cloud (e.g., Vertex AI Workbench, notebooks on Dataproc).
• Implementing security best practices in Vertex AI Workbench.
• Utilizing Spark kernels.
• Integrating with code repositories.
• Developing models in Vertex AI Workbench using common frameworks (e.g., TensorFlow, PyTorch, sklearn, Spark, JAX).

2.3 Tracking and running ML experiments. Considerations include:

• Choosing the right Google Cloud environment for development and experimentation (e.g., Vertex AI Experiments, Kubeflow Pipelines, Vertex AI TensorBoard with TensorFlow and PyTorch) based on the framework.

Section 3: Scaling prototypes into ML models 

3.1 Building models. Considerations include:

• Selecting the ML framework and model architecture.
• Applying modeling techniques based on interpretability needs.

3.2 Training models. Considerations include:

• Organizing training data (e.g., tabular, text, speech, images, videos) on Google Cloud (e.g., Cloud Storage, BigQuery).
• Ingesting various file formats (e.g., CSV, JSON, images, Hadoop, databases) for training.
• Training models using different SDKs (e.g., Vertex AI custom training, Kubeflow on Google Kubernetes Engine, AutoML, tabular workflows).
• Employing distributed training to set up robust pipelines.
• Tuning hyperparameters.
• Troubleshooting issues in ML model training.

3.3 Choosing appropriate hardware for training. Considerations include:

• Assessing compute and accelerator options (e.g., CPU, GPU, TPU, edge devices).
• Implementing distributed training with TPUs and GPUs (e.g., Reduction Server on Vertex AI, Horovod).

Section 4: Serving and scaling models 

4.1 Serving models. Considerations include:

• Implementing batch and online inference (e.g., Vertex AI, Dataflow, BigQuery ML, Dataproc).
• Serving models using various frameworks (e.g., PyTorch, XGBoost).
• Managing a model registry.
• Conducting A/B testing for different model versions.

4.2 Scaling online model serving. Considerations include:

• Utilizing Vertex AI Feature Store.
• Managing Vertex AI public and private endpoints.
• Selecting appropriate hardware (e.g., CPU, GPU, TPU, edge).
• Scaling the serving infrastructure based on throughput requirements (e.g., Vertex AI Prediction, containerized serving).
• Optimizing ML models for production in terms of performance, latency, memory, and throughput (e.g., simplification techniques).

Section 5: Automating and orchestrating ML pipelines 

5.1 Developing end-to-end ML pipelines. Considerations include:

• Validating data and models.
• Ensuring consistent data preprocessing between training and serving.
• Hosting third-party ML pipelines on Google Cloud (e.g., MLFlow).
• Identifying necessary components, parameters, triggers, and compute resources (e.g., Cloud Build, Cloud Run).
• Selecting an orchestration framework (e.g., Kubeflow Pipelines, Vertex AI Pipelines, Cloud Composer).
• Implementing hybrid or multicloud strategies.
• Designing systems using TFX components or Kubeflow DSL (e.g., Dataflow).

5.2 Automating model retraining. Considerations include:

• Defining a suitable retraining policy.
• Implementing CI/CD for model deployment (e.g., Cloud Build, Jenkins).

5.3 Tracking and auditing metadata. Considerations include:

• Tracking and comparing model artifacts and versions (e.g., Vertex AI Experiments, Vertex ML Metadata).
• Integrating with model and dataset versioning.
• Managing model and data lineage.

Section 6: Monitoring ML solutions 

6.1 Identifying risks to ML solutions. Considerations include:

• Developing safe machine learning systems (e.g., guarding against inadvertent data or model exploitation, hacking) 
• Complying with Google’s Responsible AI policies (e.g., biases)
• Evaluating the preparedness of ML solutions (e.g., fairness, data bias)
• Explainability of the Vertex AI model (Vertex AI Prediction, for example). 

6.2 Monitoring, testing, and troubleshooting ML solutions. Considerations include:

• Setting up metrics for continuous evaluation (such as Explainable AI and Vertex AI Model Monitoring) 
• Tracking training-serving skew 
• Tracking feature attribution drift 
• Tracking model performance against simpler models, baselines, and across time dimensions 
• Common training and serving errors 

Why Choose DreamsPlus?

• Expert trainers with industry experience
• Comprehensive course material
• Interactive training sessions
• Guaranteed success in Google certification exam