Ray Jin

win10下训练的模型转成ONNX后，在linux上有问题

1

Traceback (most recent call last): File "onnx/infer.py", line 119, in predictor = InferenceSession(FLAGS.onnx_file) File "/home/ljqwater/Python/lib/python3.6/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 335, in __init__ self._create_inference_session(providers, provider_options, disabled_optimizers) File "/home/ljqwater/Python/lib/python3.6/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 368, in _create_inference_session sess =...

8g显存，bs设置为4，爆显存

10

bs设为多少都爆显存

为啥输入的要求是(105,105,3),官方的是(224,224,3)啊，权重也是用得官方的，小白有点懵。

1

假如更改输入尺寸应该怎样改呢？