geoexploring

PaddleX Python代码将已训练完成模型导出为推理模型的函数接口是： ``` def _export_inference_model(self, save_dir, image_shape=None): save_dir = osp.join(save_dir, 'inference_model') self.test_inputs = self._get_test_inputs(image_shape) infer_net = self._build_inference_net() if self.status == 'Quantized': self.quantizer.save_quantized_model(infer_net, osp.join(save_dir, 'model'), self.test_inputs) quant_info = self.get_quant_info() with open(...

不支持Paddlex中的Deeplabv3+导出为ONNX模型

6

利用如下命令行进行转换： ``` paddle2onnx --model_dir C:/Users/admin/Desktop/temp --model_filename C:/Users/admin/Desktop/temp/model.pdmodel --params_filename C:/Users/admin/Desktop/temp/model.pdiparams --opset_version 11 --save_file C:/Users/admin/Desktop/temp --input_shape_dict "{'x': [1, 3, 512, 512]}" ``` 加上`--input_shape_dict "{'x': [1, 3, 512, 512]}"`报错： ``` Traceback (most recent...

defaultAllocator.cpp::nvinfer1::internal::DefaultAllocator::allocate::63] Error Code 1: Cuda Runtime (out of memory)

3

Change the input image size of `ONNX` model from `1 * 3 * 1024 * 1024` to `1 * 3 * 2000 * 2000`, get an error when `converting ONNX...

The following error is reported when I run `demo.py`: ``` (venv) C:\Users\admin\Desktop\RITM>c:/Users/admin/Desktop/RITM/venv/Scripts/python.exe c:/Users/admin/Desktop/RITM/ritm_interactive_segmentation-main/demo.py Traceback (most recent call last): File "C:\Users\admin\AppData\Local\Programs\Python\Python37\lib\distutils\spawn.py", line 69, in _spawn_nt rc = os.spawnv(os.P_WAIT, executable, cmd) FileNotFoundError:...