How ZETIC.MLange Works
Requirements and steps to start using ZETIC.MLange
Prepare model
The input for MLange is:
- Model: TorchScript or ONNX format
- Input(s): NumPy array format
For more details, refer to Prepare Model and Input(s).
Prepare On-device Model and Personal Key
Create Model Project
Use Web Dashboard or Command Line Interface to create Model Project.
Generate Personal Key
Use Web Dashboard to generate Personal Key.
For more details, refer to Generate Personal Key.
Initialize and run your model in mobile app
Please follow the Deploy to Android Studio guide for details.
val model = ZeticMLangeModel(CONTEXT, PERSONAL_KEY, PROJECT_NAME)
val output = model.run(YOUR_INPUT_TENSORS)Please follow the Deploy to Xcode guide for details.
let model = try ZeticMLangeModel(PERSONAL_KEY, PROJECT_NAME)
let output = try model.run(YOUR_INPUT_TENSORS)Profiling MLange Model
With the Web Dashboard, you can get comprehensive information about your MLange Model, including:
- Progress of generating Model Key
- Performance metrics across various devices
- Effectiveness analysis for different hardware targets