ModelAnalysis
PythonFull model analysis result.
ModelAnalysis( layers: list[LayerOps], total_macs: int, total_ops: int, num_parameters: int, engine: str = 'tflite',) -> Nonedataclass
Full model analysis result.
API tier: experimental
layers
Pythonlayers: list[LayerOps]total_macs
Pythontotal_macs: inttotal_ops
Pythontotal_ops: intnum_parameters
PythonApproximate parameter count (weights + biases).
num_parameters: intApproximate parameter count (weights + biases).
engine
PythonEngine/interpreter that produced this analysis ('tflite', 'helia-rt', 'helia-aot').
engine: str = 'tflite'Engine/interpreter that produced this analysis (‘tflite’, ‘helia-rt’, ‘helia-aot’).
ethos_u_op_count
PythonNumber of Vela-generated ethos-u custom ops in the graph.
ethos_u_op_count: intNumber of Vela-generated ethos-u custom ops in the graph.
has_ethos_u_op
PythonTrue when the model was compiled by Vela (contains ethos-u ops).
has_ethos_u_op: boolTrue when the model was compiled by Vela (contains ethos-u ops).
to_dict
PythonReturn the analysis and its layers as plain dicts.
to_dict() -> dict[str, Any]Return the analysis and its layers as plain dicts.