model.onnx file is not creating

[Igor5005]

Hello,

I ran your code and it worked and created a model for me, but only in the h5 format, the model is not created in the onnx format. Apparently, I have a weak video card, it does not support calculations... This is what happened at the end of training the model: 2025-01-31 18:26:02.160816: F tensorflow/stream_executor/cuda/cuda_driver.cc:345] Check failed: CUDA_SUCCESS == cuDevicePrimaryCtxGetState(device, &former_primary_context_flags, &former_primary_context_is_active) (0 vs. 303)

But when I tried to check the model, I ran the file, an error occurred: (venv) G:\testPyton>python inferenceModel.py Traceback (most recent call last): File "G:\testPyton\inferenceModel.py", line 31, in model = ImageToWordModel(model_path=configs.model_path, char_list=configs.vocab) File "G:\testPyton\inferenceModel.py", line 10, in init super().init(*args, **kwargs) File "G:\testPyton\venv\lib\site-packages\mltu\inferenceModel.py", line 51, in init raise Exception(f"Model path ({self.model_path}) does not exist") Exception: Model path (Models/02_captcha_to_text/202501311820\model.onnx) does not exist Please tell me how to run the model in h5 format? Otherwise I can’t do anything with it, neither check nor use it... tensorflow 2.10.0 mltu 1.2.5 onnx 1.17.0

[abTURGUT]

I'm experiencing the same issue. Can anyone who knows help? 🙂

[Igor5005]

I still haven’t been able to overcome the problem, now I’m trying to rewrite the script from Python to node.js, I understand it more...