rohit5895
/

OCR_NumInput_Base

vision-encoder-decoder

image-text-to-text

Model card Files Files and versions Community

Model Card for Model ID

Fine Tuned version of microsoft/trocr-base-stage1 on Huge number of examples of printed numerical and timestamp type values. Has a CER = 0 on hold out dataset for single line images.

Downloads last month: 68

Safetensors

Model size

334M params

Tensor type

F32

·

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support