Neural ART Compiler Error

[MaickHuguenin]

Hello,

I try to compile a model for the NUCLEO-N567 board using the STM32Cube AI Studio.

During the compilation I got an error from stedgeai core :

>>> C:\ST\STEdgeAI\4.0\Utilities\windows\stedgeai.exe analyze --model C:\Users\maick.huguenin\Downloads\Models_George\Models\model_int8.tflite --batch-size 10 --mode host --st-neural-art user_profile@C:\Users\maick.huguenin\.stm32cubeaistudio\workspace\test-N6\.ai\run\run-4\.ai\profile-neural-art.json --name network --verbosity 1 --c-api st-ai --target stm32n6 --workspace C:\Users\maick.huguenin\.stm32cubeaistudio\workspace\test-N6\.ai\run\run-4\.ai\st_ai_ws --output C:\Users\maick.huguenin\.stm32cubeaistudio\workspace\test-N6\.ai\run\run-4\.ai\st_ai_output
ST Edge AI Core v4.0.0-20500 359356bb0

 >>>> EXECUTING NEURAL ART COMPILER
   C:/ST/STEdgeAI/4.0/Utilities/windows/atonn.exe -i "C:/Users/maick.huguenin/.stm32cubeaistudio/workspace/test-N6/.ai/run/run-4/.ai/st_ai_output/model_int8_OE_3_3_1.onnx" --json-quant-file "C:/Users/maick.huguenin/.stm32cubeaistudio/workspace/test-N6/.ai/run/run-4/.ai/st_ai_output/model_int8_OE_3_3_1_Q.json" -g "network.c" --load-mdesc "C:/ST/STEdgeAI/4.0/Utilities/configs/stm32n6.mdesc" --load-mpool "C:/Users/maick.huguenin/.stm32cubeaistudio/workspace/test-N6/.ai/run/run-4/.ai/memory_pool.mpool" --load-cdesc C:/ST/STEdgeAI/4.0/Utilities/configs/cortex-m55.cdesc --save-mpool-file "C:/Users/maick.huguenin/.stm32cubeaistudio/workspace/test-N6/.ai/run/run-4/.ai/st_ai_ws/neural_art__network/memory_pool.mpool" --out-dir-prefix "C:/Users/maick.huguenin/.stm32cubeaistudio/workspace/test-N6/.ai/run/run-4/.ai/st_ai_ws/neural_art__network/" --network-name "network" --native-float --cache-maintenance --Ocache-opt --enable-virtual-mem-pools --Os --Oauto-sched --enable-epoch-controller --output-info-file "c_info.json" --generate-stai

      >>> Shell execution has FAILED (returned code = 1)
      $ C:\ST\STEdgeAI\4.0\Utilities\windows\atonn.exe -i C:\Users\maick.huguenin\.stm32cubeaistudio\workspace\test-N6\.ai\run\run-4\.ai\st_ai_output\model_int8_OE_3_3_1.onnx --json-quant-file C:\Users\maick.huguenin\.stm32cubeaistudio\workspace\test-N6\.ai\run\run-4\.ai\st_ai_output\model_int8_OE_3_3_1_Q.json -g network.c --load-mdesc C:\ST\STEdgeAI\4.0\Utilities\configs\stm32n6.mdesc --load-mpool C:\Users\maick.huguenin\.stm32cubeaistudio\workspace\test-N6\.ai\run\run-4\.ai\memory_pool.mpool --load-cdesc C:\ST\STEdgeAI\4.0\Utilities\configs\cortex-m55.cdesc --save-mpool-file C:\Users\maick.huguenin\.stm32cubeaistudio\workspace\test-N6\.ai\run\run-4\.ai\st_ai_ws\neural_art__network\memory_pool.mpool --out-dir-prefix C:\Users\maick.huguenin\.stm32cubeaistudio\workspace\test-N6\.ai\run\run-4\.ai\st_ai_ws\neural_art__network/ --network-name network --native-float --cache-maintenance --Ocache-opt --enable-virtual-mem-pools --Os --Oauto-sched --enable-epoch-controller --output-info-file c_info.json --generate-stai
      saving memory pool description "C:/Users/maick.huguenin/.stm32cubeaistudio/workspace/test-N6/.ai/run/run-4/.ai/st_ai_ws/neural_art__network/memory_pool.mpool"

      Warning: Oauto did not find valid compile options: aborting
      Configuration 0: alt-scheduler(false):
      /c/local/jenkins_cloud/workspace/2-STEDGEAI_BuildAtonnExe_Win/git/onnx_backend/src/orlando_pool_acc_v2.cc:88 pool_get_batch_size: Assertion 'i != 0' failed.
      Configuration 1: alt-scheduler(true):
      /c/local/jenkins_cloud/workspace/2-STEDGEAI_BuildAtonnExe_Win/git/onnx_backend/src/orlando_pool_acc_v2.cc:88 pool_get_batch_size: Assertion 'i != 0' failed.
      <<<
   E103(CliRuntimeError): Error calling the Neural Art compiler

Run failed: Error: analyze error: E103(CliRuntimeError): Error calling the Neural Art compiler. Please check the report for more details. Output: ST Edge AI Core v4.0.0-20500 359356bb0

PASS:   0%|          | 0/103 [00:00<?, ?it/s]
                                                       
 >>>> EXECUTING NEURAL ART COMPILER
   C:/ST/STEdgeAI/4.0/Utilities/windows/atonn.exe -i "C:/Users/maick.huguenin/.stm32cubeaistudio/workspace/test-N6/.ai/run/run-4/.ai/st_ai_output/model_int8_OE_3_3_1.onnx" --json-quant-file "C:/Users/maick.huguenin/.stm32cubeaistudio/workspace/test-N6/.ai/run/run-4/.ai/st_ai_output/model_int8_OE_3_3_1_Q.json" -g "network.c" --load-mdesc "C:/ST/STEdgeAI/4.0/Utilities/configs/stm32n6.mdesc" --load-mpool "C:/Users/maick.huguenin/.stm32cubeaistudio/workspace/test-N6/.ai/run/run-4/.ai/memory_pool.mpool" --load-cdesc C:/ST/STEdgeAI/4.0/Utilities/configs/cortex-m55.cdesc --save-mpool-file "C:/Users/maick.huguenin/.stm32cubeaistudio/workspace/test-N6/.ai/run/run-4/.ai/st_ai_ws/neural_art__network/memory_pool.mpool" --out-dir-prefix "C:/Users/maick.huguenin/.stm32cubeaistudio/workspace/test-N6/.ai/run/run-4/.ai/st_ai_ws/neural_art__network/" --network-name "network" --native-float --cache-maintenance --Ocache-opt --enable-virtual-mem-pools --Os --Oauto-sched --enable-epoch-controller --output-info-file "c_info.json" --generate-stai

                                                       
      >>> Shell execution has FAILED (returned code = 1)
      $ C:\ST\STEdgeAI\4.0\Utilities\windows\atonn.exe -i C:\Users\maick.huguenin\.stm32cubeaistudio\workspace\test-N6\.ai\run\run-4\.ai\st_ai_output\model_int8_OE_3_3_1.onnx --json-quant-file C:\Users\maick.huguenin\.stm32cubeaistudio\workspace\test-N6\.ai\run\run-4\.ai\st_ai_output\model_int8_OE_3_3_1_Q.json -g network.c --load-mdesc C:\ST\STEdgeAI\4.0\Utilities\configs\stm32n6.mdesc --load-mpool C:\Users\maick.huguenin\.stm32cubeaistudio\workspace\test-N6\.ai\run\run-4\.ai\memory_pool.mpool --load-cdesc C:\ST\STEdgeAI\4.0\Utilities\configs\cortex-m55.cdesc --save-mpool-file C:\Users\maick.huguenin\.stm32cubeaistudio\workspace\test-N6\.ai\run\run-4\.ai\st_ai_ws\neural_art__network\memory_pool.mpool --out-dir-prefix C:\Users\maick.huguenin\.stm32cubeaistudio\workspace\test-N6\.ai\run\run-4\.ai\st_ai_ws\neural_art__network/ --network-name network --native-float --cache-maintenance --Ocache-opt --enable-virtual-mem-pools --Os --Oauto-sched --enable-epoch-controller --output-info-file c_info.json --generate-stai
      saving memory pool description "C:/Users/maick.huguenin/.stm32cubeaistudio/workspace/test-N6/.ai/run/run-4/.ai/st_ai_ws/neural_art__network/memory_pool.mpool"
      Warning: Oauto did not find valid compile options: aborting
      Configuration 0: alt-scheduler(false):
      /c/local/jenkins_cloud/workspace/2-STEDGEAI_BuildAtonnExe_Win/git/onnx_backend/src/orlando_pool_acc_v2.cc:88 pool_get_batch_size: Assertion 'i != 0' failed.
      Configuration 1: alt-scheduler(true):
      /c/local/jenkins_cloud/workspace/2-STEDGEAI_BuildAtonnExe_Win/git/onnx_backend/src/orlando_pool_acc_v2.cc:88 pool_get_batch_size: Assertion 'i != 0' failed.
      <<<
   E103(CliRuntimeError): Error calling the Neural Art compiler
 

Could you help me understand this error and if it is possible to modify the model to run it on the N6 device ?

Thanks for support.

Best regards,

Maïck