Transform a machine learning model into a library for an edge device.
Transforms a trained model into a library, package, or firmware ready to deploy on an embedded device. Can optionally apply post-training quantization if a representative data sample is uploaded.
Supported model formats:
Keras Model instance <https://www.tensorflow.org/api_docs/python/tf/keras/Model>_
TensorFlow SavedModel <https://www.tensorflow.org/guide/saved_model>_ (as path to directory or .zip file)
ONNX model file <https://learn.microsoft.com/en-us/windows/ai/windows-ml/get-onnx-model>_ (as path to .onnx file)
TensorFlow Lite file <https://www.tensorflow.org/lite/guide>_ (as bytes, or path to any file that is not .zip or .onnx)
Representative data for quantization:
Must be a numpy array or .npy file.
Each element must have the same shape as your model's input.
Must be representative of the range (maximum and minimum) of values in your training data.
Note: the available deployment options will change depending on the values given for model, model_output_type, and model_input_type. For example, the openmv deployment option is only available if model_input_type is set to ImageInput. If you attempt to deploy to an unavailable target, you will receive the error Could not deploy: deploy_target: ....
Parameters
model: Union[pathlib.Path, str, bytes, Any]
model_output_type: Union[edgeimpulse.model.output_type.Classification, edgeimpulse.model.output_type.Regression, edgeimpulse.model.output_type.ObjectDetection]
model_input_type: Union[edgeimpulse.model.input_type.ImageInput, edgeimpulse.model.input_type.AudioInput, edgeimpulse.model.input_type.TimeSeriesInput, edgeimpulse.model.input_type.OtherInput, ForwardRef(None)] = None
representative_data_for_quantization: Union[pathlib.Path, str, bytes, Any, ForwardRef(None)] = None
deploy_model_type: Optional[str] = None
engine: str = 'tflite'
deploy_target: str = 'zip'
output_directory: Optional[str] = None
api_key: Optional[str] = None
timeout_sec: Optional[float] = None
Return
_io.BytesIO
List suitable deployment targets for the project associated with configured or provided api key.
Parameters
api_key: Optional[str] = None
Return
List[str]
List all the engines that can be passed to deploy()'s engine parameter.
Returns: List[str]: List of engines
Return
List[str]
List all the model types that can passed to deploy()'s deploy_model_type parameter.
Returns: List[str]: List of model types
Return
List[str]
List possible values for the device field when calling edgeimpulse.model.profile().
Parameters
api_key: Optional[str] = None
Return
List[str]
Profile the performance of a trained model on a range of embedded targets, or a specific device.
The response includes estimates of memory usage and latency for the model across a range of targets, including low-end MCU, high-end MCU, high-end MCU with accelerator, microprocessor unit (MPU), and a GPU or neural network accelerator. It will also include details of any conditions that preclude operation on a given type of device.
If you request a specific device, the results will also include estimates for that specific device. A list of devices can be obtained from edgeimpulse.model.list_profile_devices().
You can call .summary() on the response to obtain a more readable version of the most relevant information.
Parameters
model: Union[pathlib.Path, str, bytes, Any]
device: Optional[str] = None
api_key: Optional[str] = None
timeout_sec: Optional[float] = None
Return
edgeimpulse.model._functions.profile.ProfileResponse