Lacax
/

deepseek_ocr_lora

@@ -1,207 +1,112 @@
 ---
 tags:
 - unsloth
 ---
----
-library_name: transformers
-  pipeline_tag: image-text-to-text
-  tags:
-    - transformers
-    - safetensors
-    - unsloth
-    - endpoints_compatible
-# Model Card for Model ID
-<!-- Provide a quick summary of what the model is/does. -->
-## Model Details
-### Model Description
-<!-- Provide a longer summary of what this model is. -->
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
-### Model Sources [optional]
-<!-- Provide the basic links for the model. -->
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
-## Uses
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-### Direct Use
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-[More Information Needed]
-### Downstream Use [optional]
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-[More Information Needed]
-### Out-of-Scope Use
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-[More Information Needed]
-## Bias, Risks, and Limitations
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-[More Information Needed]
-### Recommendations
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
-## How to Get Started with the Model
-Use the code below to get started with the model.
-[More Information Needed]
-## Training Details
-### Training Data
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-[More Information Needed]
-### Training Procedure
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-#### Preprocessing [optional]
-[More Information Needed]
-#### Training Hyperparameters
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-#### Speeds, Sizes, Times [optional]
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-[More Information Needed]
-## Evaluation
-<!-- This section describes the evaluation protocols and provides the results. -->
-### Testing Data, Factors & Metrics
-#### Testing Data
-<!-- This should link to a Dataset Card if possible. -->
-[More Information Needed]
-#### Factors
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-[More Information Needed]
-#### Metrics
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-[More Information Needed]
-### Results
-[More Information Needed]
-#### Summary
-## Model Examination [optional]
-<!-- Relevant interpretability work for the model goes here -->
-[More Information Needed]
-## Environmental Impact
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
-## Technical Specifications [optional]
-### Model Architecture and Objective
-[More Information Needed]
-### Compute Infrastructure
-[More Information Needed]
-#### Hardware
-[More Information Needed]
-#### Software
-[More Information Needed]
-## Citation [optional]
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
-**BibTeX:**
-[More Information Needed]
-**APA:**
-[More Information Needed]
-## Glossary [optional]
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
-[More Information Needed]
-## More Information [optional]
-[More Information Needed]
-## Model Card Authors [optional]
-[More Information Needed]
-## Model Card Contact
-[More Information Needed]

 ---
+language:
+- es
+license: apache-2.0
+base_model: unsloth/DeepSeek-OCR-2
 tags:
+- ocr
+- receipt
+- invoice
+- vision
+- fine-tuned
 - unsloth
+- lora
+- json-extraction
+pipeline_tag: image-text-to-text
 ---
+# DeepSeek OCR LoRA — Extracción estructurada de tickets y facturas
+## Descripción
+Este modelo es un **LoRA fine-tuning** sobre [`unsloth/DeepSeek-OCR-2`](https://huggingface.co/unsloth/DeepSeek-OCR-2), entrenado para analizar imágenes de **tickets de compra, recibos y facturas** y devolver su contenido en formato **JSON estructurado**.
+El objetivo principal del proyecto fue académico: estudiar el comportamiento de un modelo de visión multimodal al ser adaptado para extraer información con una estructura de salida fija.
+El modelo está diseñado principalmente para documentos en **español**.
+---
+## Salida esperada
+Dado una imagen de ticket o factura, el modelo devuelve un JSON con la siguiente estructura:
+```json
+{
+  "comercio": "Nombre del establecimiento",
+  "fecha": "DD/MM/AAAA",
+  "cif": "B12345678",
+  "productos": [
+    {
+      "producto": "Nombre del producto",
+      "cantidad": 2,
+      "precio": 3.50
+    }
+  ],
+  "precio_total": 7.00
+}
+```
+---
+## Uso
+Este modelo requiere cargar el modelo base por separado y aplicar el LoRA encima:
+```python
+from transformers import AutoModelForCausalLM, AutoProcessor
+from peft import PeftModel
+BASE_MODEL_ID = "unsloth/DeepSeek-OCR-2"
+LORA_ID       = "Lacax/deepseek_ocr_lora"
+processor = AutoProcessor.from_pretrained(BASE_MODEL_ID)
+model = AutoModelForCausalLM.from_pretrained(BASE_MODEL_ID)
+model = PeftModel.from_pretrained(model, LORA_ID)
+```
+> **Nota:** El repo del LoRA es privado. Se requiere un `HF_TOKEN` con acceso al repo `Lacax/deepseek_ocr_lora`. El modelo base `unsloth/DeepSeek-OCR-2` es público.
+---
+## Datos de entrenamiento
+- **Tipo:** Imágenes de tickets y facturas reales (fotografías propias)
+- **Tamaño:** ~100 imágenes
+- **Formato:** Cada imagen asociada a su JSON ground truth con la estructura descrita arriba
+- **Idioma:** Español
+- **Origen:** Dataset propio, no publicado
+---
+## Estadísticas de entrenamiento
+| Parámetro              | Valor       |
+|------------------------|-------------|
+| Épocas completadas     | 3           |
+| Duración total         | ~4607 s (~77 min) |
+| Muestras / segundo     | 0.40        |
+| Pérdida final (train)  | 0.0399      |
+| Framework              | Unsloth + PEFT |
+| Plataforma             | RunPod (GPU cloud) |
+---
+## Limitaciones
+- Dataset pequeño (~100 imágenes): el modelo puede no generalizar bien a formatos de ticket muy distintos a los usados en entrenamiento.
+- Entrenado únicamente con documentos en español; rendimiento no garantizado en otros idiomas.
+- Proyecto de carácter académico/experimental, no validado en producción.
+- La pérdida baja de entrenamiento (0.04) puede indicar cierto sobreajuste al dataset propio.
+---
+## Modelo base
+- [`unsloth/DeepSeek-OCR-2`](https://huggingface.co/unsloth/DeepSeek-OCR-2)
+---
+## Autor
+**Lacax** — Proyecto de estudio personal sobre fine-tuning de modelos de visión multimodal.