wscang

I used model generated by the following instruction. **python3 tools/pytorch2onnx.py configs/top_down/hrnet/coco/hrnet_w48_coco_256x192.py https://download.openmmlab.com/mmpose/top_down/hrnet/hrnet_w48_coco_256x192-b9e0b3ab_20200708.pth --output-file hrnet_w48_coco_256x192.onnx**

I tried the following codes. **import onnxruntime import numpy as np** **sess_options = onnxruntime.SessionOptions() sess = onnxruntime.InferenceSession('hrnet_w48_coco_256x192.onnx', sess_options) data = [np.random.rand(1, 3, 256, 192).astype(np.float32)] input_names = sess.get_inputs() feed = zip(sorted(i_.name...

The problem has been solved. Using a different version of TensorRT was a problem. This is my mistake. Thank you.

Can you provide onnx model file with Google Drive or OneDrive? Thank you.