# distilbert-command-data-tagger
`distilbert-command-data-tagger` is a lightweight, optimized ONNX model trained for joint **Intent Classification** and **Named Entity Recognition (NER) / Data Tagging** from natural language user commands.
It is designed for lightweight deployment on low-resource runtimes and Hugging Face Inference Endpoints using ONNX Runtime.
---
## Model Architecture & Files
- **Base Model**: `distilbert-base-uncased`
- **Model Format**: ONNX (`joint_command_parser.onnx` + `joint_command_parser.onnx.data`)
- **Max Sequence Length**: 64 tokens
- **Output Heads**:
1. **Intent Logits**: Decodes top-level command intent (with a strict `0.65` confidence threshold fallback to `unknown`).
2. **NER / Data Tag Logits**: Tagging token sequences (e.g., target text to edit, recipient email address, or search queries).
---
## Supported Intents
The model categorizes inputs into one of the following structured categories:
- **Document Editing**: `add_text`, `search_d…
