ModalityType
Enums · AiDotNet.Data.Multimodal
Identifies the type of data modality in a multimodal sample.
For Beginners
A “modality” means a type of data. Modern AI models often work with multiple types at once - for example, an image captioning model uses both images and text. This enum identifies which type each piece of data is.
Fields
| Field | Summary |
|---|---|
ModalityType Audio |
Audio data (waveform samples or spectrograms). |
ModalityType Custom |
Custom or unspecified modality type. |
ModalityType Graph |
Graph-structured data (nodes and edges). |
ModalityType Image |
Image data (2D pixel grids, typically [H, W, C] or [C, H, W]). |
ModalityType PointCloud |
3D point cloud data (sets of 3D coordinates with optional features). |
ModalityType Tabular |
Structured/tabular data (feature vectors). |
ModalityType Text |
Text data (token sequences, embeddings, or raw strings). |
ModalityType TimeSeries |
Time series data (sequential measurements over time). |
ModalityType Video |
Video data (sequences of frames, typically [T, H, W, C]). |
int value__ |