search

TUI-NICR / ESANet

ESANet: Efficient RGB-D Semantic Segmentation for Indoor Scene Analysis
Other
231 stars 49 forks source link

The results of evaluation #13

Open Huhaowen0130 opened 3 years ago

Huhaowen0130 commented 3 years ago

Hello, I have run the evaluation process: "python eval.py \ --dataset sunrgbd \ --dataset_dir ./datasets/sunrgbd \ --ckpt_path ./trained_models/sunrgbd/r34_NBt1D.pth", where are the results saved?

mona0809 commented 3 years ago

The evaluation script simply computes mIoU and prints it to console.

danielS91 commented 3 years ago

However, for qualitative results, you can still use inference_dataset.py (see: inference section in our readme). If you want to store the evaluations results combining both inference_dataset.py and eval.py might be interesting for you.

Huhaowen0130 commented 3 years ago

The evaluation script simply computes mIoU and prints it to console.

OK, thank you!

Huhaowen0130 commented 3 years ago

However, for qualitative results, you can still use inference_dataset.py (see: inference section in our readme). If you want to store the evaluations results combining both inference_dataset.py and eval.py might be interesting for you.

Thank you, I'll try!