Floating point exception (core dumped)

[nguyenhd]

Hi,

I exported ONNX to RKNN but I am getting this warning and error:

/home/nguyenhd2/.local/lib/python3.5/site-packages/rknn/api/rknn.py:150: RuntimeWarning: overflow encountered in exp ret = self.rknn_base.load_onnx(model, inputs, input_size_list, outputs) W RKNN do not support type float64, try convert to float32 W RKNN do not support type float64, try convert to float32 W RKNN do not support type float64, try convert to float32 W RKNN do not support type float64, try convert to float32 .... W Input channel 1 less then 3, please update NPU Driver to v1.3.0 or later. W:tensorflow:From /home/nguyenhd2/.local/lib/python3.5/site-packages/rknn/api/rknn.py:150: The name tf.ConfigProto is deprecated. Please use tf.compat.v1.ConfigProto instead.

W:tensorflow:From /home/nguyenhd2/.local/lib/python3.5/site-packages/rknn/api/rknn.py:150: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead.

W:tensorflow:From /home/nguyenhd2/.local/lib/python3.5/site-packages/rknn/api/rknn.py:150: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.

W Warning: Axis may need to be adjusted according to original model shape. W Warning: Axis may need to be adjusted according to original model shape. W Warning: Axis may need to be adjusted according to original model shape. W Warning: Axis may need to be adjusted according to original model shape. .... done --> Building model done --> Export RKNN model done

---

Target platforms filled in RKNN model: ['RK1808'] Target platforms supported by this RKNN model: ['RK1808', 'RK1806', 'RK3399PRO']

---

--> Init runtime environment librknn_runtime version 1.7.1 (bd41dbc build: 2021-10-28 16:15:23 base: 1131) Floating point exception (core dumped)

[zen-xingle]

Hello, would you mind uploading the model? Then we can reproduce this problem.

[nguyenhd]

Hello, would you mind uploading the model? Then we can reproduce this problem.

Thank you! I have fixed the problem. The issue comes from my model

[QianLingjun]

What's wrong with the model, and how do I correct the model? thinks