44 lines
1.5 KiB
Markdown
44 lines
1.5 KiB
Markdown
---
|
||
language: en
|
||
license: apache-2.0
|
||
tags:
|
||
- home-assistant
|
||
- voice-assistant
|
||
- automation
|
||
- assistant
|
||
- home
|
||
pipeline_tag: text-generation
|
||
datasets:
|
||
- acon96/Home-Assistant-Requests
|
||
base_model:
|
||
- TinyLlama/TinyLlama-1.1B-Chat-v1.0
|
||
base_model_relation: finetune
|
||
---
|
||
|
||
# 🏠 TinyLLaMA-1.1B Home Assistant Voice Model
|
||
|
||
This model is a **fine-tuned version** of [TinyLlama/TinyLlama-1.1B-Chat-v1.0](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0), trained with [acon96/Home-Assistant-Requests](https://huggingface.co/datasets/acon96/Home-Assistant-Requests).
|
||
It is designed to act as a **voice-controlled smart home assistant** that takes natural language instructions and outputs **Home Assistant commands**.
|
||
|
||
---
|
||
|
||
## ✨ Features
|
||
- Converts **natural language voice commands** into Home Assistant automation calls.
|
||
- Produces **friendly confirmations** and **structured JSON service commands**.
|
||
- Lightweight (1.1B parameters) – runs efficiently on CPUs, GPUs, and via **Ollama** with quantization.
|
||
|
||
---
|
||
|
||
## 🔧 Example Usage (Transformers)
|
||
|
||
```python
|
||
from transformers import AutoModelForCausalLM, AutoTokenizer
|
||
|
||
tokenizer = AutoTokenizer.from_pretrained("premrajreddy/tinyllama-1.1b-home-llm")
|
||
model = AutoModelForCausalLM.from_pretrained("premrajreddy/tinyllama-1.1b-home-llm")
|
||
|
||
query = "turn on the kitchen lights"
|
||
inputs = tokenizer(query, return_tensors="pt")
|
||
outputs = model.generate(**inputs, max_new_tokens=80)
|
||
|
||
print(tokenizer.decode(outputs[0], skip_special_tokens=True)) |