Open saurabhtangri opened 1 month ago
Describe the bug A clear and concise description of what the bug is.
To Reproduce Steps to reproduce the behavior:
SSD-MobilenetV1-12-qdq
import onnx_tool import numpy as np modelpath = '/home/ssd_mobilenet_v1_13-qdq.onnx' m = onnx_tool.Model(modelpath) m.graph.shape_infer({'inputs': np.ones((1,383,640,3), dtype=np.uint8)}) # update tensor shapes with new input tensor m.graph.profile() m.graph.print_node_map() # console print
m.save_model('/home/ssd_mobilenet_v1_13-qdq-mods.onnx', shape_only=False) # only with weight tensor shapes and dynamic tensor shapes
Screenshots If applicable, add screenshots to help explain your problem.
it's a known issue that Loop is not supported. But this model is not simplified. The real model body does not have this Loop node. It should start with this Mul node.
Describe the bug A clear and concise description of what the bug is.
To Reproduce Steps to reproduce the behavior:
SSD-MobilenetV1-12-qdq
import onnx_tool import numpy as np modelpath = '/home/ssd_mobilenet_v1_13-qdq.onnx' m = onnx_tool.Model(modelpath) m.graph.shape_infer({'inputs': np.ones((1,383,640,3), dtype=np.uint8)}) # update tensor shapes with new input tensor m.graph.profile() m.graph.print_node_map() # console print
m.save_model('/home/ssd_mobilenet_v1_13-qdq-mods.onnx', shape_only=False) # only with weight tensor shapes and dynamic tensor shapes
Screenshots If applicable, add screenshots to help explain your problem.