Arize AI
Search…
⌃K

Model Types

How to pick a performance metric and map your model schema based on model type

Model Types Overview

Machine learning models predict different outcomes based on various model parameters. These model types determine the data ingestion format and the performance metrics that can be visualized in the platform.

Performance Metrics by Model Type

Your model type determines your performance metric, follow this chart to map performance metrics relevant to your model type.
Metrics are batched into Metric Families that align with model types and their variants.
Metric Family
Classification
Metrics
Accuracy, Recall, Precision, FPR, FNR, F1, Sensitivity, Specificity
Metric Family
Regression
Metrics
MAPE, MAE, RMSE, MSE, R2, Mean Error
Metric Family
Ranking
Metric Family
AUC/Log Loss
Metrics
AUC, PR-AUC, Log Loss
Metric Family
Computer Vision
Metrics
MAP, IoU
For more information on how these are defined or calculated, please refer to Model Metric Definitions

Ranking

Variant
NLP Classification
Performance Metric
Classification Metrics
Variant
NLP NER
Performance Metric
Classification Metrics
Variant
NLP POS Tagging
Performance Metric
Classification Metrics
Variant
CV Classification
Performance Metric
Classification Metrics
Variant
CV Object Detection
Performance Metric
Computer Vision Metrics
Variant
CV Instance Segmentation
Performance Metric
Computer Vision Metrics
Classification Metrics
Variant
CV Semantic/Pixel Segmentation
Performance Metric
Computer Vision Metrics
Classification Metrics