Model development
Model development is the process of designing, constructing, training, evaluating, and refining a computational or statistical model for a defined task.
Delivering Machine Learning Projects: From design to deployment
Machine learning modeling is the process of selecting, training, debugging, tuning, and refining machine learning models from data.
- How it is used
- Turns understood data and application constraints into designed, tracked, and evaluated models.
- How deep it goes
- Specifies process tickets, design decisions, tracking, and evaluation risks, but does not provide a training algorithm walkthrough.
- Examples
- Uses a smart-building time-series split to show how apparently strong test performance can mislead.
- What you can do after
- Practice/perform a documented modeling process that tracks experiments, tests performance, and rejects unsuitable models.
Data Without Labels
A model development process is an organized sequence of activities for defining a problem, preparing data, developing a model, and preparing it for deployment.
- How it is used
- Organizes machine-learning work from business problem definition through model deployment and maintenance.
- How deep it goes
- Traces the full lifecycle and details data preparation, EDA, iterative algorithm selection, approval, deployment, documentation, and maintenance.
- Examples
- Uses measurable business targets, cleaned datasets, exploratory reports, candidate algorithms, hyperparameter iterations, performance measures, and stakeholder feedback.
- What you can do after
- Apply it to select, approve, deploy, document, and maintain a model that addresses a measurable business problem.
Designing Deep Learning Systems: A guide for software engineers
The process of designing, training, and validating a machine learning model.
- How it is used
- Translates a deep learning use case into data preparation, algorithm selection, training, and inference validation.
- How deep it goes
- Defines the end-to-end modeling workflow and production handoff, but stops short of designing the supporting deep learning system.
- Examples
- Uses notebooks, dataset management services, packaged training and inference code, and deployment strategies as workflow artifacts.
- What you can do after
- Build and validate trained models, package their code, and prepare satisfactory inference for product integration.
Related concepts
- data flow diagramincludes · A data flow diagram is a visual model of how data moves among processes, components, data stores, and external sources or destinations in a system.
- algorithmproduces · An algorithm is a finite, precisely specified procedure for performing a computation, transforming inputs into outputs, or solving a problem.
- feature engineeringhas part · Feature engineering is the creation, transformation, or selection of input features from data to provide useful representations for analysis or modeling.
- TensorFlowused by · TensorFlow is an open-source software platform for numerical computation and machine learning, with tools for building, training, and deploying models.
- machine learning workflowpart of · A machine learning workflow is an organized sequence of activities for formulating, developing, evaluating, deploying, operating, and maintaining a machine learning system.
- language modelproduces · A language model is a computational model that represents patterns in linguistic tokens to estimate sequence probabilities and predict or generate language from context.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.