Enable type constraint check for node inputs and outputs in onnx checker. Today a model with unsupported types for an operator does not fail checker. Adding such a check strengthens the checker and enables for thorough validation of the model.Example: This model should fail checker. < ir_vers...
onnx.checker.check_model(model) return model@@ -311,7 +295,7 @@ def check_and_update_input_shapes(model: onnx.ModelProto, input_shapes: TensorShdef simplify(model: Union[str, onnx.ModelProto], check_n: int = 0, perform_optimization: bool = True, ...
for (const bool is_sce_internal : {true, false}) { auto pre_graph_checker = [is_sce_internal](Graph& graph) -> Status { auto op_count_pre = CountOpsInGraph(graph); TEST_RETURN_IF_NOT(op_count_pre.size() == 1U); if (is_sce_internal) TEST_RETURN_IF_NOT(op_count_pre["com....
onnx: 1.14.0 opencensus: 0.11.2 opencensus-context: 0.1.3 opencv-contrib-python: 4.8.0.74 opencv-python: 4.8.0.74 opencv-python-headless: 4.8.0.74 openpyxl: 3.1.2 openslide-python: 1.2.0 opentelemetry-api: 1.18.0 opentelemetry-exporter-otlp: 1.18.0 opentelemetry-exporter-otlp-proto-common: 1....
self.assertFalse(inc_config.quantization["is_static"]) num_quantized_matmul = 0 for initializer in onnx_model.graph.initializer: if "MatMul" in initializer.name and "quantized" in initializer.name: num_quantized_matmul += 1 num_quantized_matmul = num_quantized_matmul_onnx_model(onnx_model)...