Skip to main content
Version: 0.10.0

Type Alias: PaddleOcrModel

PaddleOcrModel = object

Defined in: extensions/cv/tasks/paddleOcr.ts:91

Model configuration for the PP-OCRv6 pipeline: one fused detect/recognize PTE, the charset published beside it, and the run options.

Properties

charsetPath

readonly charsetPath: string

Defined in: extensions/cv/tasks/paddleOcr.ts:99

The recognizer charset published beside the model: a JSON array of strings, one per class, where charset[i] labels logit i + 1 (logit 0 is the CTC blank). Resolved to a local path by the resource fetcher like modelPath.


modelOpts

readonly modelOpts: PaddleOcrModelOptions

Defined in: extensions/cv/tasks/paddleOcr.ts:101

Run options. See PaddleOcrModelOptions.


modelPath

readonly modelPath: string

Defined in: extensions/cv/tasks/paddleOcr.ts:93

The fused detect/recognize PTE. Resolved to a local path by the fetcher.