Enabling software components: the machine learning model utilizes, relies on and consists of software frameworks, libraries, and runtime environments necessary for its development, training and operation. These components consist of necessary tools for building, training, evaluating, management and deployment of machine learning models. Example software frameworks: TensorFlow, Keras and Pytorch. For LLM’s, Hugging Face Transforms resource provides a collection of pre-trained models, tokenizers and other components utilized for LLM development. The can be many software components “*”, which are utilized for development and maintenance of one model.
Defined methods of the system asset:
- createMLModel() – a method representing the process of building the initial Preprocessed machine learning model.
- deployMLModel() – method defining a process of deployment and start of utilization of Deployed, trained and tuned model.
- processQueryInputData() – after model’s input ingests user’s input data through intermediary APIs and services, this method preprocesses it. The preprocessing may involve necessary tokenization, filtering, normalization to ensure desired quality of model’s inferences.
Involved business asset within the association with the machine learning model:
- Machine learning model development process: a systematic continuous process of machine learning development, incorporating software components together to create a model, enable its training and conduct inference operations. The goal is to build a machine learning model, which will be able to solve the target business specific problem to achieve a target objective. The process may involve activities like data collection, data preprocessing, model design, building and training, quality assurance, deployment, monitoring, feedback, and auditability.