Instructions to use arnabwithab/laya-android-control with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Laya
How to use arnabwithab/laya-android-control with Laya:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
laya-android-control
Finetune of convaiinnovations/laya for Android phone control.
Training data
- Source: eros438/phone-decisions-ac.
- That file holds 18,354 decision rows.
- The rows convert xwm/AndroidControl.
- Training ran 4 epochs on a Kaggle T4 with
laya-traindefaults.
Results
| Model | test operations accuracy |
|---|---|
| Base Laya-english | 0.261 |
| Head-only finetune | 0.574 |
| Full finetune | 0.665 |
In-sample calibration gave accuracy 0.85 and ECE 0.037.
Use
import laya
agent = laya.load("arnabwithab/laya-android-control", device="cuda")
agent.predict("Goal: ...", {"operation": {"type": "choice",
"instructions": "Which operation moves this phone task forward?",
"criteria": {"TAP": "...", "TYPE": "...", "SCROLL": "...",
"BACK": "...", "WAIT": "...", "DONE": "..."}}})
Serve it with laya-serve pointed at this repo.
- Downloads last month
- 33
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for arnabwithab/laya-android-control
Base model
convaiinnovations/laya