WHAT YOU WILL LEARN
Put together knowledge, detect statistical knowledge biases, carry out function engineering at scale to coach fashions, & practice, consider, & tune fashions with AutoML
Retailer & handle ML options utilizing a function retailer, & debug, profile, tune, & consider fashions whereas monitoring knowledge lineage and mannequin artifacts
Construct, deploy, monitor, & operationalize end-to-end machine studying pipelines
Construct knowledge labeling and human-in-the-loop pipelines to enhance mannequin efficiency with human intelligence
SKILLS YOU WILL GAIN
- Pure Language Processing with BERT
- ML Pipelines and ML Operations (MLOps)
- A/B Testing and Mannequin Deployment
- Information Labeling at Scale
- Automated Machine Studying (AutoML)
- Statistical Information Bias Detection
- Multi-class Classification with FastText and BlazingText
- Information ingestion
- Exploratory Information Evaluation
- ML Pipelines and MLOps
- Mannequin Coaching and Deployment with BERT
- Mannequin Debugging and Analysis
About this Specialization
Growth environments may not have the precise necessities as manufacturing environments. Transferring knowledge science and machine studying initiatives from thought to manufacturing requires state-of-the-art abilities. You must architect and implement your initiatives for scale and operational effectivity. Information science is an interdisciplinary subject that mixes area data with arithmetic, statistics, knowledge visualization, and programming abilities.
The Sensible Information Science Specialization brings collectively these disciplines utilizing purpose-built ML instruments within the AWS cloud. It helps you develop the sensible abilities to successfully deploy your knowledge science initiatives and overcome challenges at every step of the ML workflow utilizing Amazon SageMaker.
This Specialization is designed for data-focused builders, scientists, and analysts conversant in the Python and SQL programming languages who need to discover ways to construct, practice, and deploy scalable, end-to-end ML pipelines – each automated and human-in-the-loop – within the AWS cloud.
Every of the ten weeks contains a complete lab developed particularly for this Specialization that gives hands-on expertise with state-of-the-art algorithms for pure language processing (NLP) and pure language understanding (NLU), together with BERT and FastText utilizing Amazon SageMaker.
Utilized Studying Challenge
By the tip of this Specialization, you’ll be able to:
• Ingest, register, and discover datasets
• Detect statistical bias in a dataset
• Routinely practice and choose fashions with AutoML
• Create machine studying options from uncooked knowledge
• Save and handle options in a function retailer
• Prepare and consider fashions utilizing built-in algorithms and customized BERT fashions
• Debug, profile, and evaluate fashions to enhance efficiency
• Construct and run an entire ML pipeline end-to-end
• Optimize mannequin efficiency utilizing hyperparameter tuning
• Deploy and monitor fashions
• Carry out knowledge labeling at scale
• Construct a human-in-the-loop pipeline to enhance mannequin efficiency
• Cut back value and enhance efficiency of knowledge merchandise
0 Comments