MLOps Wiki

Data Integrity

Data integrity refers to reliable and accurate data with consistency and context for confident decisions and improved business agility.

Data Pipeline

A data pipeline includes a set of processes and tools to combine, automate, compute, and transform data from a source to a destination.

Data Segments

Data segments help effectively analyze data by dividing it and grouping similar data together according to the chosen parameter or filter.

Training-Serving Skew

Training-serving skew is a gap in the ML model’s performance during training and serving due to data changes and handling discrepancies.

Concept Drift

A type of model drift observed due to the changes in the properties of dependent variables or the target to be predicted over time.

Data Drift

A shift in the distribution of features in training and serving data while the relationship between the input and target is unchanged.

Model Drift

The degradation of the ML model’s predictive power as a result of changes in the digital environment and variables like data and concepts.

ML Monitors

ML monitors are tools that help check model performance, identify prediction quality issues, and generate real-time alerts to inform users.

Model Behavior

Model behavior is explained by global or local interpretations and corresponds to features, target predictions, and learning from segments.

Model Fairness

ML Fairness is the model’s quality or state of being fair or impartial, and it relates to the harm of allocation and quality of services.

Predictions

Prediction is an information output generated by an ML algorithm trained on historical input data.

CI/CD for Machine Learning

CI/CD of ML pipelines enables teams to build source code, run tests, and deploy automated pipelines for continuous delivery and training.

Data And Model Versioning

Versioning is the process of uniquely naming multiple iterations of models used at different stages of ML development to track all changes.

Feature Management

A feature store is a central repository to store features by defining a feature transformation once and computing its values to serve models.

LLMOps

LLMOps is an evolving subfield of MLOps that focuses on operationalizing large language models(LLMs) at scale.

ML Diagnostics

ML diagnostics include tests to identify potential issues and apply improvements at different stages of the ML development process.

ML Model Metrics

ML model metrics help to evaluate the performance of the machine learning models.

ML Reproducibility

ML reproducibility is the ability to replicate the ML workflow previously carried out and produce the same results as the original work.

ML Scalability

ML scalability is scaling ML models to handle massive data sets and perform many computations in a cost-effective and time-saving way.

ML Stack

An ML stack is a reference model listing all infrastructural components required to build, deploy, and scale machine learning systems.

ML governance

ML Governance is an internal framework to control the processes for ML development, implement and track activities, and assign roles.

MLOps Tools

MLOps tools simplify the complex ML development process and enable maintainability and auditability with ML experiments.

ModelOps

ModelOps is a holistic approach to enable the smooth operationalization of ML models to deliver expected business value to an enterprise.

Pipeline and Process Automation

Automated ML streamlines the complete ML pipeline - data extraction, processing, and storing for scalability, efficiency, and productivity.

AI Artifacts

AI Artifacts denote the output generated by the training process. They are produced at various stages and used across the ML lifecycle.

Ground Truth

In the ML context, ground truth denotes factual data that is observed or computed and can be analyzed objectively.

MLOps

MLOps defines practices that unify ML development, streamline continuous delivery of models, and enable collaboration between teams

Machine Learning

Machine learning allows computers to learn and act like humans without being explicitly programmed and improves their learning over time

Machine Learning Lifecycle

The ML lifecycle defines sequential steps involved in data science projects that carry equal importance and are executed in a cyclic order

Machine Learning Model

An ML model is an object trained over a set of data to learn from it, identify patterns, and infer over data that has not been seen before

Model Monitoring

ML monitoring is the practice of tracking the performance of ML models in production to identify potential issues in ML pipelines

Natural Language Processing (NLP)

NLP, or Natural Language Processing, is a field in Artificial Intelligence that allows computers to understand and comprehend the languages that are written by Humans.

Overfitting Vs. Underfitting

Model overfitting and underfitting scenarios depict how well the model learns the training data. Finding a good fit between these two is imperative.

AI Audit

AI audit defines a piloted audit program to educate C-suite leaders, expose risks involved, and accordingly develop safeguard controls

AI Black Box

Black box AI refers to using ML models that are not explainable by looking at their parameters, making retracing ML outputs difficult

AI Explainability

AI explainability refers to a set of processes that allows humans to understand, comprehend and trust the results produced by ML algorithms

AI Observability

AI observability is a holistic and complete approach to drive insights on the model’s behavior, data, and performance across its lifecycle.

Bias In Machine Learning

ML bias is the effect of erroneous assumptions used in the processes or prejudices in the data producing systematically prejudiced results

Ethical AI

Ethical AI defines guidelines for individual rights, behavior manipulation, privacy, and non-discrimination to ensure legitimate use of AI

Responsible AI

Responsible AI (RAI) brings practices to develop, deploy, and scale AI for good causes that impact people and society fairly