Insight Foundation - Introduction
AI Workbench toolsets are to enable IDEA users to perform all stages of AI/ML activities. IDEA platform intends to support the following AI/ML activities
- Annotator Ground Truth: To enable labelling structured/tabular datasets for ML model training as well as varieties of unstructured datasets for Cognitive AI model training.
- AI Build: To experiment: To enable experimenting with data for preparation, cleaning and feature engineering to generate data ready for AI/ML model training.
- AI Train: To enable training AI/ML models on training ready datasets
- AI Deploy: To enable deploying available AI/ML models for live inference/scoring or support batch inference/scoring
- AI Monitor: To enable monitoring deployed AI/ML models for detecting Data Drift or Model Drift
- AutoML: To support naïve or early AI/ML users through encapsulated and automated AI Build and AI Train
- AI Trust: To enable trust and transparency of AI Train instances through advanced insights
- Reproducible AI MLOps Pipeline: To enable DevOps of AI/ML pipelines by stitching together artefacts of core building blocks, i.e., Build, Train, Deploy, Monitor and Trust through CI/CD
- • The intent of AI Workbench toolsets is to derive predictive insights and prescriptive actions on data available in AI datastore. It assumes that required data sources have already been ingested through appropriate transformations and these ingested data are available in Data Lake, say ADLS Gen2 (refer to below high level design diagram of IDEA platform)
The intent of AI Workbench toolsets is to derive predictive insights and prescriptive actions on data available in AI datastore. It assumes that required data sources have already been ingested through appropriate transformations and these ingested data are available in Data Lake, say ADLS Gen2 (refer to below high level design diagram of IDEA platform).
IDEA AI Workbench attempts to solve the myriads of prevalent inconveniences throughout AI/ML lifecycles faced by client teams of Data Scientists and Machine Learning.
