Upload README.md with huggingface_hub

README.md

@@ -0,0 +1,125 @@
+---
+tags:
+- bert
+- transformers
+- litert
+- tflite
+- edge
+- on-device
+license: mit
+base_model: FacebookAI/roberta-base
+pipeline_tag: feature-extraction
+---
+
+# roberta-base - LiteRT
+
+This is a [LiteRT](https://ai.google.dev/edge/litert) (formerly TensorFlow Lite) conversion of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) for efficient on-device inference.
+
+## Model Details
+
+| Property | Value |
+|----------|-------|
+| **Original Model** | [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) |
+| **Format** | LiteRT (.tflite) |
+| **File Size** | 473.7 MB |
+| **Task** | Feature Extraction / Classification Base |
+| **Max Sequence Length** | 128 |
+| **Output Dimension** | 768 |
+| **Pooling Mode** | N/A (Full hidden states) |
+
+## Performance
+
+Benchmarked on Intel CPU (WSL2):
+
+| Metric | Value |
+|--------|-------|
+| **Inference Latency** | 81.2 ms |
+| **Throughput** | 12.3/sec |
+| **Cosine Similarity vs Original** | 1.0000 ✅ |
+
+## Quick Start
+
+```python
+import numpy as np
+from ai_edge_litert.interpreter import Interpreter
+from transformers import AutoTokenizer
+
+# Load model and tokenizer
+interpreter = Interpreter(model_path="FacebookAI_roberta-base.tflite")
+interpreter.allocate_tensors()
+input_details = interpreter.get_input_details()
+output_details = interpreter.get_output_details()
+
+tokenizer = AutoTokenizer.from_pretrained("FacebookAI/roberta-base")
+
+def get_hidden_states(text: str) -> np.ndarray:
+    """Get hidden states for input text."""
+    encoded = tokenizer(
+        text,
+        padding="max_length",
+        max_length=128,
+        truncation=True,
+        return_tensors="np"
+    )
+
+    interpreter.set_tensor(input_details[0]["index"], encoded["input_ids"].astype(np.int64))
+    interpreter.set_tensor(input_details[1]["index"], encoded["attention_mask"].astype(np.int64))
+    interpreter.invoke()
+
+    return interpreter.get_tensor(output_details[0]["index"])
+
+# Example
+hidden = get_hidden_states("Hello, world!")
+cls_embedding = hidden[0, 0, :]  # CLS token for classification
+print(f"Hidden shape: {hidden.shape}")  # (1, 128, 768)
+```
+
+## Files
+
+- `FacebookAI_roberta-base.tflite` - The LiteRT model file
+
+## Conversion Details
+
+- **Conversion Tool**: [ai-edge-torch](https://github.com/google-ai-edge/ai-edge-torch)
+- **Conversion Date**: 2026-01-12
+- **Source Framework**: PyTorch → LiteRT
+- **Validation**: Cosine similarity 1.0000 vs original
+
+## Intended Use
+
+- **Mobile Applications**: On-device semantic search, RAG systems
+- **Edge Devices**: IoT, embedded systems, Raspberry Pi
+- **Offline Processing**: Privacy-preserving inference
+- **Low-latency Applications**: Real-time processing
+
+## Limitations
+
+- Fixed sequence length (128 tokens)
+- CPU inference (GPU delegate requires setup)
+- Tokenizer loaded separately from original model
+- Float32 precision
+
+## License
+
+This model inherits the license from the original:
+- **License**: MIT ([source](https://huggingface.co/FacebookAI/roberta-base))
+
+## Citation
+
+```bibtex
+@article{liu2019roberta,
+    title={RoBERTa: A Robustly Optimized BERT Pretraining Approach},
+    author={Liu, Yinhan and Ott, Myle and others},
+    journal={arXiv preprint arXiv:1907.11692},
+    year={2019}
+}
+```
+
+## Acknowledgments
+
+- Original model by [FacebookAI](https://huggingface.co/FacebookAI)
+- Conversion using [ai-edge-torch](https://github.com/google-ai-edge/ai-edge-torch)
+
+---
+
+*Converted by [Bombek1](https://huggingface.co/Bombek1)*