Skip to main content

Hi,

I get an error when I want to deploy the WASB model for sports ball detection and tracking, converted to onnx with opset 15:

ERROR   : Traceback (most recent call last):
ERROR   :   File "<frozen compiler.top_level>", line 584, in quantize
ERROR   :   File "<frozen qtools_tvm_interface.graph_exporter_v2.graph_exporter>", line 254, in export
ERROR   :   File "<frozen qtools_tvm_interface.graph_exporter_v2.graph_exporter>", line 209, in _convert_operators
ERROR   :   File "<frozen qtools_tvm_interface.graph_exporter_v2.replacement_functions>", line 242, in closure
ERROR   : RuntimeError: Multiple consecutive arithmetic operations by runtime ops found in the graph. This is not allowed presently.
ERROR   : 
ERROR   : The above exception was the direct cause of the following exception:
ERROR   : 
ERROR   : Traceback (most recent call last):
ERROR   :   File "/home/aetina/Desktop/voyager-sdk/axelera/app/compile.py", line 456, in compile
ERROR   :     the_manifest = top_level.compile(model, compilation_cfg, output_path)
ERROR   :   File "<frozen compiler.utils.error_report>", line 65, in wrapper
ERROR   :   File "<frozen compiler.top_level>", line 831, in compile
ERROR   :   File "<frozen compiler.utils.error_report>", line 65, in wrapper
ERROR   :   File "<frozen compiler.top_level>", line 586, in quantize
ERROR   : axelera.compiler.exceptions.GraphExporterError: Quantized graph simplification failed:
ERROR   : Multiple consecutive arithmetic operations by runtime ops found in the graph. This is not allowed presently.
ERROR   : Quantized graph simplification failed:
ERROR   : Multiple consecutive arithmetic operations by runtime ops found in the graph. This is not allowed presently.
INFO    : Quantizing wasb-onnx: wasb-onnx took 1097.626 seconds
ERROR   : Failed to deploy network
 

Hey ​@Giodst! This could be due to multiple consecutive arithmetic operations, that Voyager can’t simplify during quantisation, perhaps.

Are you able to re-export the model using ONNX opset 14? I’m wondering if that might have better compatibility.

https://github.com/axelera-ai-hub/voyager-sdk/blob/release/v1.3/docs/tutorials/custom_model.md


Hi ​@Spanner,

I tried with opset 14 but I get the same error.

Thanks.


Reply