Recent Posts
Zero-shot time series forecasting with Chronos using Amazon Bedrock and ClickHouse
Large Language Models
Amazon Chronos
Amazon Bedrock
The emergence of large language models (LLMs) with zero-shot generalization capabilities in sequence modelling tasks has led to the development of time series foundation models (TSFMs) based on LLM architectures. By representing time series as sequences of tokens, TSFMs can leverage LLMs’ capability to extrapolate future patterns from the context data. TSFMs eliminate the traditional need for domain-specific model development, allowing organizations to deploy accurate time series solutions faster.
Forecasting inflation with AutoML in Amazon SageMaker Autopilot
Amazon SageMaker
Machine Learning
Automated Machine Learning (AutoML) frameworks address the expertise- and time-intensive nature of the traditional ML model development workflow. By automating the end-to-end process of building, training and tuning ML models through easy-to-use programmatic interfaces, AutoML solutions make ML accessible to non-specialists and significantly reduce deployment timeframes.