TexasInstruments / edgeai-tidl-tools

Edgeai TIDL Tools and Examples - This repository contains Tools and example developed for Deep learning runtime (DLRT) offering provided by TI’s edge AI solutions.
Other
121 stars 27 forks source link

[ONNXRuntimeError] : 1 : FAIL : This is an invalid model. Error: the graph is not acyclic. #38

Open aneesh1514 opened 1 year ago

aneesh1514 commented 1 year ago

I am trying to compile the mobilenet-SSD based model on onnxruntime. But I am facing an error.

Preliminary subgraphs created = 2 Final number of subgraphs created are : 1, - Offloaded Nodes - 48, Total Nodes - 55 Node in deny list...delegated to ARM --- layer type - Transpose, Node name - Transpose_45 Node in deny list...delegated to ARM --- layer type - Reshape, Node name - Reshape_46 Node in deny list...delegated to ARM --- layer type - Transpose, Node name - Transpose_50 Node in deny list...delegated to ARM --- layer type - Reshape, Node name - Reshape_51 Node in deny list...delegated to ARM --- layer type - Concat, Node name - Concat_52 Node in deny list...delegated to ARM --- layer type - Concat, Node name - Concat_53 Node in deny list...delegated to ARM --- layer type - Softmax, Node name - Softmax_54 Running runtimes graphviz - /data/home/paneesh/edgeai-tidl-tools/tidl_tools/tidl_graphVisualiser_runtimes.out ../../../model-artifacts//fd//allowedNode.txt ../../../model-artifacts//fd//tempDir/graphvizInfo.txt ../../../model-artifacts//fd//tempDir/runtimes_visualization.svg Process Process-1: Traceback (most recent call last): File "/usr/lib/python3.6/multiprocessing/process.py", line 258, in _bootstrap self.run() File "/usr/lib/python3.6/multiprocessing/process.py", line 93, in run self._target(*self._args, self._kwargs) File "onnxrt_ep.py", line 174, in run_model sess = rt.InferenceSession(config['model_path'] ,providers=EP_list, provider_options=[delegate_options, {}], sess_options=so) File "/data/home/paneesh/.local/lib/python3.6/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 283, in init self._create_inference_session(providers, provider_options) File "/data/home/paneesh/.local/lib/python3.6/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 315, in _create_inference_session sess.initialize_session(providers, provider_options) onnxruntime.capi.onnxruntime_pybind11_state.Fail: [ONNXRuntimeError] : 1 : FAIL : This is an invalid model. Error: the graph is not acyclic.**

The onnx model is converted from a PyTorch model.

But the model is working with the CPU.(CPUEXECUTIONPROVIDER).

Please help me in solving the issue.

mash2612 commented 1 year ago

@aneesh1514 were you able to resolve this issue?

aneesh1514 commented 1 year ago

No, Still we are not able to solve this issue.