Update model card with training info

README.md

@@ -1,135 +1,79 @@
 ---
-title: Myanmar Ghost
-emoji: 🇲🇲
-colorFrom: blue
-colorTo: purple
-sdk: gradio
-sdk_version: 4.0.0
-app_file: app.py
-pinned: false
+language:
+- my
+- en
+license: apache-2.0
+library_name: transformers
+pipeline_tag: text-classification
+tags:
+- myanmar
+- burmese
+- nlp
+- text-classification
+- sentiment-analysis
+- news-classification
+- xlm-roberta
 ---
 
-# Myanmar Ghost
+# Myanmar Ghost 🐉
 
-Advanced Myanmar Language Understanding Model with Multi-Modal Fusion
+Advanced Myanmar Language Understanding Model
 
 ## 🎯 Overview
 
-Myanmar Ghost is a comprehensive NLP project for Myanmar (Burmese) language understanding, featuring:
+Myanmar Ghost is a transformer-based language model designed for Myanmar (Burmese) language understanding tasks.
 
-- **Multi-Modal Data Fusion**: Combines audio prosody with text for sentiment analysis
-- **Active Learning Pipeline**: Efficient annotation with uncertainty sampling
-- **Federated Learning Ready**: Privacy-preserving distributed training
-- **Explainable AI**: SHAP and LIME-based model interpretability
-- **Adversarial Data Augmentation**: Robust training with synonym replacement and perturbation
+### Features
 
-## 📁 Project Structure
-
-```
-Myanmar-Ghost/
-├── configs/              # Configuration files
-├── data/                 # Data directory (raw, processed, augmented)
-├── src/
-│   ├── data_processing/  # Audio, text processing, multimodal fusion
-│   ├── annotation/        # Labeling tools, active learning
-│   ├── models/           # Model implementations
-│   ├── federated/        # Federated learning components
-│   ├── xai/              # Explainable AI modules
-│   ├── augmentation/      # Data augmentation
-│   └── utils/            # Utilities
-├── outputs/              # Model outputs, results, logs
-├── tests/                # Unit tests
-└── docs/                 # Documentation
-```
-
-## 🚀 Quick Start
-
-### Installation
-
-```bash
-pip install -r requirements.txt
-```
-
-### Data Processing
-
-```bash
-python src/data_processing/audio_processor.py
-python src/data_processing/text_normalizer.py
-```
-
-### Training
-
-```bash
-python src/models/train.py --config configs/model/transformer_config.yaml
-```
-
-### Federated Learning
-
-```bash
-# Start server
-python src/federated/server.py --config configs/federated/server_config.yaml
-
-# Start client
-python src/federated/client.py --config configs/federated/client_configs/hospital_a.yaml
-```
-
-## 📊 Features
-
-### 1. Data Versioning (DVC)
+- **Multi-task Learning**: Supports classification, translation, and NLI
+- **Pre-trained on Myanmar corpus**: 10.6M Myanmar text samples
+- **Fine-tuned datasets**: News classification, translation, instruction tuning
+- **Multi-modal Ready**: Audio + text fusion for sentiment analysis
 
-Track data changes with DVC for reproducibility.
-
-```bash
-dvc add data/processed/v1.2
-git add data/processed/v1.2.dvc
-```
+## 📊 Training Data
 
-### 2. Active Learning
+| Dataset | Size | Purpose |
+|---------|------|---------|
+| Myanmar Written Corpus | 10.6M | Pre-training |
+| Myanmar News Classification | 29K | Fine-tuning |
+| Myanmar-English Translation | 22K | Translation |
+| Instruction Tuning | 91K | Chat/QA |
 
-Select uncertain samples for annotation:
+## 🚀 Quick Start
 
 ```python
-from src.annotation.active_learning.uncertainty_sampler import UncertaintySampler
-
-sampler = UncertaintySampler(model, method="least_confidence")
-selected = sampler.select_samples(samples, tokenizer, n_samples=100)
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+
+# Load model
+model_name = "amkyawdev/myanmar-ghost"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForSequenceClassification.from_pretrained(model_name)
+
+# Predict
+text = "မြန်မာစာသတင်း"
+inputs = tokenizer(text, return_tensors="pt")
+outputs = model(**inputs)
+prediction = outputs.logits.argmax(dim=-1)
 ```
 
-### 3. XAI Explanations
-
-Understand model decisions:
-
-```python
-from src.xai.shap_explainer import ThankingSHAPExplainer
+## 📁 Project Structure
 
-explainer = ThankingSHAPExplainer(model, tokenizer)
-shap_values = explainer.explain("ကျေးဇူးပါ")
 ```
-
-### 4. Data Augmentation
-
-```python
-from src.augmentation.synonym_replacer import MyanmarSynonymReplacer
-from src.augmentation.perturbator import TextPerturbator
-
-replacer = MyanmarSynonymReplacer()
-aug_text = replacer.augment_text("ကျေးဇူးပါ", replace_prob=0.3)
+Myanmar-Ghost/
+├── configs/           # Model & training configs
+├── data/             # Datasets
+├── src/
+│   ├── models/       # Model implementations
+│   ├── data_processing/
+│   ├── training/     # Training pipeline
+│   └── evaluation/   # Metrics & benchmarking
+└── docs/             # Documentation
 ```
 
-## 📈 CI/CD
-
-GitHub Actions workflows automatically:
-
-- Validate data schemas
-- Check label distributions
-- Run model evaluations
-- Deploy to HuggingFace
-
-## 📝 License
+## 📜 License
 
-Apache 2.0 License
+Apache 2.0
 
 ## 👤 Author
 
-Aung Myo Kyaw (amkyawdev)  
-https://huggingface.co/amkyawdev
+Aung Myo Kyaw (amkyawdev)