Update README.md

README.md

@@ -1,102 +1,102 @@
----
-license: gpl-3.0
-task_categories:
-- text-generation
-language:
-- en
-tags:
-- chemistry
-- molecular-editing
-- drug-discovery
-- smiles
-- molecule-generation
-pretty_name: MEGA Molecular Editing Dataset (522K)
-size_categories:
-- 100K<n<1M
----
-
-# MEGA: A Large-Scale Molecular Editing Dataset for Guided-Action Optimization
-
-Large-scale molecular editing dataset with 522K examples for training models to modify molecular structures based on natural language instructions.
-
-**Paper**: [MEGA: A Large-Scale Molecular Editing Dataset for Guided-Action Optimization](https://openreview.net/pdf?id=MaS7e2EVFm)
-**Official Repository**: [https://github.com/nfsrules/MEGA-moledit](https://github.com/nfsrules/MEGA-moledit)
-
-## Usage
-
-```python
-from datasets import load_dataset
-
-# Load dataset
-dataset = load_dataset("nfsrulesFR/mega-moledit-522K")
-
-# Access splits
-# Positive examples (successful edits)
-train_data = dataset["train"]
-val_data = dataset["validation"]
-# Negative examples (unsuccessful edits)
-train_neg_data = dataset["train_neg"]
-val_neg_data = dataset["validation_neg"]
-
-# Example
-example = train_data[0]
-print(f"Prompt: {example['prompt']}")
-print(f"Input: {example['input_smiles']}")
-print(f"Output: {example['output_smiles']}")
-```
-
-## Supported Tasks
-
-| Task ID | Description |
-|---------|-------------|
-| 101 | Increase water solubility |
-| 102 | Decrease water solubility |
-| 103 | Increase drug-likeness |
-| 104 | Decrease drug-likeness |
-| 105 | Increase permeability |
-| 106 | Decrease permeability |
-| 107 | Increase hydrogen bond acceptors |
-| 108 | Increase hydrogen bond donors |
-| 201 | Increase solubility + HBA |
-| 202 | Decrease solubility + increase HBA |
-| 203 | Increase solubility + HBD |
-| 204 | Decrease solubility + increase HBD |
-| 205 | Increase solubility + permeability |
-| 206 | Increase solubility + decrease permeability |
-
-## Dataset Structure
-
-Each example contains:
-- `task_id`: Task identifier
-- `prompt`: Natural language instruction
-- `input_smiles`: Input molecule
-- `output_smiles`: Target molecule
-- `action_type`: Edit operation type
-- `edit`: Specific edit applied
-- `target_delta`: Change in target property
-- `SA_delta`: Change in Synthetic Accessibility
-- `MW_delta`: Change in Molecular Weight
-- `QED_delta`: Change in Drug-likeness
-- `murcko_scaffold_retained`: Scaffold preservation flag
-
-**Splits**: `train` (469K), `validation` (52K), `train_neg` (469K), `validation_neg` (52K)
-
-## Trained Models
-
-Llama 3 8B-based models for molecular optimization:
-
-- **MEGA-SFT**: [nfsrulesFR/mega-sft](https://huggingface.co/nfsrulesFR/mega-sft) - Supervised fine-tuning model
-- **MEGA-GRPO**: [nfsrulesFR/mega-grpo](https://huggingface.co/nfsrulesFR/mega-grpo) - Tanimoto-GRPO optimized model
-
-## Citation
-
-```bibtex
-@article{
-  fernandez2025mega,
-  title={MEGA: A Large-Scale Molecular Editing Dataset for Guided-Action Optimization},
-  author={Nelson Fernandez and Maxime Illouz and Luis Pinto and Entao Yang and Habiboulaye Amadou Boubacar},
-  journal={Under review at International Conference on Learning Representations},
-  year={2025},
-  url={https://openreview.net/pdf?id=MaS7e2EVFm}
-}
-```
+---
+license: gpl-3.0
+task_categories:
+- text-generation
+language:
+- en
+tags:
+- chemistry
+- molecular-editing
+- drug-discovery
+- smiles
+- molecule-generation
+pretty_name: MEGA Molecular Editing Dataset (522K)
+size_categories:
+- 100K<n<1M
+---
+
+# MEGA: A Large-Scale Molecular Editing Dataset for Guided-Action Optimization
+
+Large-scale molecular editing dataset with 522K examples for training models to modify molecular structures based on natural language instructions.
+
+**Paper**: [MEGA: A Large-Scale Molecular Editing Dataset for Guided-Action Optimization](https://openreview.net/pdf?id=MaS7e2EVFm)
+**Official Repository**: [https://github.com/nfsrules/MEGA-moledit](https://github.com/nfsrules/MEGA-moledit)
+
+## Usage
+
+```python
+from datasets import load_dataset
+
+# Load dataset
+dataset = load_dataset("nfsrulesFR/mega-moledit-522K")
+
+# Access splits
+# Positive examples (successful edits)
+train_data = dataset["train"]
+val_data = dataset["validation"]
+# Negative examples (unsuccessful edits)
+train_neg_data = dataset["train_neg"]
+val_neg_data = dataset["validation_neg"]
+
+# Example
+example = train_data[0]
+print(f"Prompt: {example['prompt']}")
+print(f"Input: {example['input_smiles']}")
+print(f"Output: {example['output_smiles']}")
+```
+
+## Supported Tasks
+
+| Task ID | Description |
+|---------|-------------|
+| 101 | Increase water solubility |
+| 102 | Decrease water solubility |
+| 103 | Increase drug-likeness |
+| 104 | Decrease drug-likeness |
+| 105 | Increase permeability |
+| 106 | Decrease permeability |
+| 107 | Increase hydrogen bond acceptors |
+| 108 | Increase hydrogen bond donors |
+| 201 | Increase solubility + HBA |
+| 202 | Decrease solubility + increase HBA |
+| 203 | Increase solubility + HBD |
+| 204 | Decrease solubility + increase HBD |
+| 205 | Increase solubility + permeability |
+| 206 | Increase solubility + decrease permeability |
+
+## Dataset Structure
+
+Each example contains:
+- `task_id`: Task identifier
+- `prompt`: Natural language instruction
+- `input_smiles`: Input molecule
+- `output_smiles`: Target molecule
+- `action_type`: Edit operation type
+- `edit`: Specific edit applied
+- `target_delta`: Change in target property
+- `SA_delta`: Change in Synthetic Accessibility
+- `MW_delta`: Change in Molecular Weight
+- `QED_delta`: Change in Drug-likeness
+- `murcko_scaffold_retained`: Scaffold preservation flag
+
+**Splits**: `train` (469K), `validation` (52K), `train_neg` (469K), `validation_neg` (52K)
+
+## Trained Models
+
+Llama 3 8B-based models for molecular optimization:
+
+- **MEGA-SFT**: [nfsrulesFR/mega-sft](https://huggingface.co/nfsrulesFR/mega-sft) - Supervised fine-tuning model
+- **MEGA-GRPO**: [nfsrulesFR/mega-grpo](https://huggingface.co/nfsrulesFR/mega-grpo) - Tanimoto-GRPO optimized model
+
+## Citation
+
+```bibtex
+@article{
+  fernandezillouz2025mega,
+  title={MEGA: A Large-Scale Molecular Editing Dataset for Guided-Action Optimization},
+  author={Nelson Fernandez and Maxime Illouz and Luis Pinto and Entao Yang and Habiboulaye Amadou Boubacar},
+  journal={Under review at International Conference on Learning Representations},
+  year={2025},
+  url={https://openreview.net/pdf?id=MaS7e2EVFm}
+}
+```