Upload README.md

README.md

@@ -1,6 +1,5 @@
 ---
 title: "MoMa: Modular Deep Learning Framework for Material Property Prediction"
-emoji: 🧪
 colorFrom: blue
 colorTo: purple
 sdk: static
@@ -14,14 +13,12 @@ tags:
 - pytorch
 ---
 
-# 🧪 MoMa: Modular Deep Learning Framework for Material Property Prediction
+# MoMa: Modular Deep Learning Framework for Material Property Prediction
 
 <div align="center">
 
 [![arXiv](https://img.shields.io/badge/arXiv-2502.15483-b31b1b.svg)](https://arxiv.org/abs/2502.15483)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
-[![Python 3.8+](https://img.shields.io/badge/python-3.8+-blue.svg)](https://www.python.org/downloads/)
-[![PyTorch](https://img.shields.io/badge/PyTorch-2.0+-ee4c2c.svg)](https://pytorch.org/)
 [![HuggingFace Models](https://img.shields.io/badge/🤗-Models-yellow)](https://huggingface.co/yuanhangtangle-air/moma-modules)
 
 **[📄 Paper](https://arxiv.org/abs/2502.15483) | [🌐 Project Page](https://yuanhangtangle-air.github.io/moma-modules/) | [💻 Code](https://github.com/your-repo) | [📊 Datasets](https://huggingface.co/datasets/your-datasets)**
@@ -55,14 +52,14 @@ Our repository contains **107 pre-trained modules** spanning **6 major domains**
 
 | Domain | Count | Description |
 |--------|-------|-------------|
-| **🔬 Electronic Structure** | 28 | Band gaps, HOMO-LUMO, DOS, dielectric properties |
-| **🌡️ Thermodynamics** | 20 | Formation energy, Gibbs free energy, stability |
-| **⚗️ Spectroscopy** | 24 | EXAFS, XANES spectral features |
-| **🧲 Mechanical** | 8 | Elastic moduli, piezoelectric properties |
-| **🔋 Photovoltaic** | 8 | Solar cell performance metrics |
-| **🌊 Adsorption** | 8 | Gas adsorption in MOFs |
-| **⚡ Thermoelectric** | 8 | Seebeck coefficients, thermal conductivity |
-| **🔍 Other Properties** | 3 | Specialized material characteristics |
+| ** Electronic Structure** | 28 | Band gaps, HOMO-LUMO, DOS, dielectric properties |
+| ** Thermodynamics** | 20 | Formation energy, Gibbs free energy, stability |
+| ** Spectroscopy** | 24 | EXAFS, XANES spectral features |
+| ** Mechanical** | 8 | Elastic moduli, piezoelectric properties |
+| ** Photovoltaic** | 8 | Solar cell performance metrics |
+| ** Adsorption** | 8 | Gas adsorption in MOFs |
+| **Thermoelectric** | 8 | Seebeck coefficients, thermal conductivity |
+| ** Other Properties** | 3 | Specialized material characteristics |
 
 ### 🔍 Featured Modules
 
@@ -104,24 +101,13 @@ Our repository contains **107 pre-trained modules** spanning **6 major domains**
 
 </details>
 
-## 🛠️ Installation & Usage
 
-### Quick Start
-
-```bash
-# Install required packages
-pip install torch huggingface_hub pandas tqdm
-
-# Clone or download the modules
-git clone https://huggingface.co/yuanhangtangle-air/moma-modules
-```
-
-### 📥 Download Individual Modules
+## 📥 Download Individual Modules
 
 ```python
 import shutil
 from huggingface_hub import hf_hub_download
-
+# see all module names at https://huggingface.co/yuanhangtangle-air/moma-modules/blob/main/module-names.csv
 # Download a specific module
 cached_file = hf_hub_download(
     repo_id="yuanhangtangle-air/moma-modules",
@@ -134,7 +120,7 @@ save_path = "./moma-hub/"
 shutil.copy(cached_file, save_path)
 ```
 
-### 📦 Download All Modules
+## 📦 Download All Modules
 
 ```python
 import shutil
@@ -142,93 +128,31 @@ import pandas as pd
 from huggingface_hub import hf_hub_download
 from tqdm import tqdm
 
-# Load module list
-module_info_url = "https://huggingface.co/yuanhangtangle-air/moma-modules/resolve/main/domain-dataset-concept-en.csv"
-df = pd.read_csv(module_info_url)
-module_list = df['Dataset']
+# fetch 
+cached_file = hf_hub_download(
+    repo_id="yuanhangtangle-air/moma-modules",
+    filename="module-names.csv",
+    repo_type="model"
+)
+df = pd.read_csv(cached_file)
 
-# Download all modules
+MODULE_LIST = df['module_names']
 cached_files = []
-for module_name in tqdm(module_list, desc="Downloading modules"):
+for module_name in tqdm(MODULE_LIST, desc="Downloading Modules"):
     cached_file = hf_hub_download(
-        repo_id="yuanhangtangle-air/moma-modules",
-        filename=f"{module_name}.pt",
+    repo_id="yuanhangtangle-air/moma-modules",
+        filename=module_name + ".pt",
         repo_type="model"
     )
     cached_files.append(cached_file)
 
-# Copy to local directory
+# save to local path
 save_path = "./moma-hub/"
 for cached_file in cached_files:
     shutil.copy(cached_file, save_path)
 ```
 
-### 🚀 Using MoMa Framework
-
-```python
-# Import the framework (pseudo-code - actual implementation may vary)
-from moma import MaterialPropertyPredictor, ModularNetwork
-
-# Initialize with pre-trained modules
-model = MaterialPropertyPredictor(
-    network=ModularNetwork.from_pretrained('material-property-base'),
-    modules=['structure_encoder', 'property_decoder']
-)
-
-# Load your material data
-materials = load_material_dataset('your_dataset.csv')
-
-# Train the model
-model.train(materials, epochs=100, batch_size=32)
-
-# Make predictions
-predictions = model.predict(new_materials)
-
-# Evaluate performance
-metrics = model.evaluate(test_materials)
-print(f"Accuracy: {metrics['accuracy']:.3f}")
-```
-
-## 📊 Performance Results
-
-Our experimental evaluation across **17 datasets** demonstrates MoMa's superior performance:
-
-- 🥇 **Rank 1.35** average ranking among 5 competing methods
-- 🏆 **14/17 tasks** achieve state-of-the-art performance  
-- 📈 **14% average improvement** over strongest baseline
-- ⚡ **Superior few-shot learning** capabilities
-- 🔄 **Excellent continual learning** performance
-
-## 📁 Module File Structure
-
-```
-moma-modules/
-├── README.md                           # This file
-├── domain-dataset-concept-en.csv       # Module descriptions
-├── Electronic_Structure/               # 28 modules
-│   ├── HL_Gap.pt
-│   ├── HOMO_Energy.pt
-│   ├── LUMO_Energy.pt
-│   └── ...
-├── Thermodynamics/                     # 20 modules
-│   ├── jarvis_eform.pt
-│   ├── gibbs_free_energy.pt
-│   └── ...
-├── Spectroscopy/                       # 24 modules
-│   ├── spec_EXAFS_Fe_EdgeEnergy.pt
-│   ├── spec_XANES_Co_PeakHeight.pt
-│   └── ...
-└── Other_Domains/                      # 35 modules
-    ├── Mechanical/
-    ├── Photovoltaic/
-    ├── Adsorption/
-    ├── Thermoelectric/
-    └── ...
-```
-
-## 🔬 Research Impact
-
-### 📝 Citation
+## 📝 Citation
 
 If you use MoMa in your research, please cite our paper:
 
@@ -241,31 +165,17 @@ If you use MoMa in your research, please cite our paper:
 }
 ```
 
-### 🌟 Key Contributions
-
-1. **Novel Modular Paradigm**: First framework to replace pre-training with specialized module composition
-2. **Comprehensive Evaluation**: Extensive benchmarking across 17 diverse material property datasets
-3. **Open Science**: All 107 modules freely available to accelerate materials discovery
-4. **Superior Performance**: Consistent improvements over traditional approaches
-
-## 🤝 Community & Support
-
-- 📧 **Contact**: [Your contact information]
-- 🐛 **Issues**: [Report bugs or request features]
-- 💬 **Discussions**: [Community discussions]
-- 📚 **Documentation**: [Detailed documentation]
-
 ## 📜 License
 
-This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
+This project is licensed under the MIT License.
 
 ## 🙏 Acknowledgments
 
 We gratefully acknowledge the materials science community and the datasets that made this research possible. Special thanks to:
 
-- Materials Project (MP)
-- Open Quantum Materials Database (OQMD)
-- Joint Automated Repository for Various Integrated Simulations (JARVIS)
+- [Materials Project](https://legacy.materialsproject.org/)
+- [Open Quantum Materials Database](https://oqmd.org/materials/)
+- [Joint Automated Repository for Various Integrated Simulations](https://jarvis.nist.gov/)
 - And all other data contributors
 
 ---