Saltar al contenido
ModelPublicadoExterno
Descargar

Nyansapo Ocr Trocr LoRA v1 0 1

maxtargetlen reduced from 128 → 4 (root fix for hallucination) norepeatngramsize set to 0 (was hurting short outputs) lengthpenalty set to 1.0 (was 2.0 — was pushing model to generate more tokens)

TipoLoRAModelo basemicrosoft/trocr-base-handwrittenVersiónadapter_modelCcalidad

Fuente del modelo

Tipo
LoRA
Tamaño del archivo
19,2 MB
Modelo base
microsoft/trocr-base-handwritten
Versión
vadapter_model
Actualizado
18 sept 2026
Descargas
2
Fuente
Hugging Face

Descripción de la fuente

Nyansapo OCR — TrOCR LoRA v1.0.1

What changed from v1.0.0

  • max_target_len reduced from 128 → 4 (root fix for hallucination)
  • no_repeat_ngram_size set to 0 (was hurting short outputs)
  • length_penalty set to 1.0 (was 2.0 — was pushing model to generate more tokens)

Training details

  • Base model: microsoft/trocr-base-handwritten
  • PEFT: LoRA (r=16, alpha=32)
  • Dataset: Nzyoka19/ocr_images

Fuentes

1 fuente
Hugging FacePrincipalVersión 16de768bb493279e5711d554447f005b39903e39
huggingface.co/Nzyoka19/nyansapo-ocr-trocr-lora_v1-0-1/tree/16de768bb493279e5711d554447f005b39903e39

Verificado 3 sept

Artefactos del modelo

1 artefacto

adapter_model.safetensors

safetensors · 19,2 MB · SHA-256 2acc18493828…b285 · Hugging Face

Download file

Extractos de fuentes

2 extractos

--- base_model: microsoft/trocr-base-handwritten tags: [ocr, lora, peft, trocr, image-to-text, nyansapo] license: mit --- # Nyansapo OCR — TrOCR LoRA v1.0.1 ## What changed from v1.0.0 - `max_target_len` reduced from 128 → 4 (root fix for hallucination) - `no_repeat_ngram_size` set to 0 (was hurting short outputs) - `length_penalty` set to 1.0 (was 2.0 — was pushing model to generate more tokens) ## Training details - **Base model:** microsoft/trocr-base-handwritten - **PEFT:** LoRA (r=16, alpha=32) - **Dataset:** Nzyoka19/ocr_images

Source context: 2 downloads · 0 likes · Pipeline image-to-text · Library peft · Repo Nzyoka19/nyansapo-ocr-trocr-lora_v1-0-1