Upload README.md with huggingface_hub

README.md

@@ -1,67 +1,73 @@
 ---
 library_name: peft
 license: apache-2.0
-base_model: answerdotai/ModernBERT-base
 tags:
-- base_model:adapter:answerdotai/ModernBERT-base
+- json-extraction
+- modernbert
 - lora
-- transformers
-model-index:
-- name: DiffuBERTa
-  results: []
+- diffuberta
+language: en
+metrics:
+- name: train_loss
+  value: 4.7773
+- name: eval_loss
+  value: 4.318033695220947
+datasets:
+- generated-json-pairs
 ---
 
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
-
-# DiffuBERTa
-
-This model is a fine-tuned version of [answerdotai/ModernBERT-base](https://huggingface.co/answerdotai/ModernBERT-base) on an unknown dataset.
-It achieves the following results on the evaluation set:
-- Loss: 4.3180
-
-## Model description
-
-More information needed
-
-## Intended uses & limitations
-
-More information needed
-
-## Training and evaluation data
+---
+datasets:
+- generated-json-pairs
+language: en
+library_name: peft
+license: apache-2.0
+metrics:
+- name: train_loss
+  value: 4.7773
+- name: eval_loss
+  value: 4.318033695220947
+tags:
+- json-extraction
+- modernbert
+- lora
+- diffuberta
+---
 
-More information needed
+# DiffuBERTa: JSON Extraction Adapter
 
-## Training procedure
+This model is a Fine-tuned version of **answerdotai/ModernBERT-base** using LoRA. It is designed to extract structured JSON data from unstructured text using a parallel decoding approach.
 
-### Training hyperparameters
+## Model Performance
+- **Final Training Loss**: 4.7773
+- **Final Evaluation Loss**: 4.318033695220947
+- **Training Epochs**: 5
+- **Date Trained**: 2025-11-28
 
-The following hyperparameters were used during training:
-- learning_rate: 3e-05
-- train_batch_size: 8
-- eval_batch_size: 8
-- seed: 42
-- gradient_accumulation_steps: 2
-- total_train_batch_size: 16
-- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
-- lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 500
-- num_epochs: 5
+## 🚀 Live Demo Output
+*(Generated automatically after training)*
 
-### Training results
+**Input Text:**
+> "We are excited to welcome Dr. Sarah to our Paris office as Senior Data Scientist."
 
-| Training Loss | Epoch | Step | Validation Loss |
-|:-------------:|:-----:|:----:|:---------------:|
-| 15.3944       | 1.0   | 63   | 14.6510         |
-| 13.7016       | 2.0   | 126  | 10.3954         |
-| 10.2371       | 3.0   | 189  | 6.2723          |
-| 5.5815        | 4.0   | 252  | 5.0812          |
-| 4.7773        | 5.0   | 315  | 4.3180          |
+**Template:**
+> `{'name': '[1]', 'job': '[2]', 'city': '[1]'}`
 
+**Model Output:**
+```json
+{
+  "name": "Sarah",
+  "job": "Data scientist",
+  "city": "Paris"
+}
+```
 
-### Framework versions
+## Usage
+```python
+from transformers import AutoModelForMaskedLM, AutoTokenizer
+from peft import PeftModel
 
-- PEFT 0.18.0
-- Transformers 4.57.3
-- Pytorch 2.9.1+cu128
-- Tokenizers 0.22.1
+base_model = AutoModelForMaskedLM.from_pretrained("answerdotai/ModernBERT-base")
+model = PeftModel.from_pretrained(base_model, "philipp-zettl/DiffuBERTa")
+# ... use extract_parallel helper ...
+```