the ./dataset/SAMUS/CAMUS_full/class.json is missing

[jasonrichdarmawan]

Commit:

36f1b98

Goal:

run $ python test_video.py

Issues:

• the medpy package is not in the requirements.txt

  Error log

  (.venv) jason@Jasons-MacBook-Pro 02-MemSAM % python test_video.py
  Traceback (most recent call last):
    File "/Users/jason/01 Learn/Active/Programming/03 NPU/03 Large Model/02-MemSAM/test_video.py", line 20, in <module>
      from utils.evaluation import get_eval
    File "/Users/jason/01 Learn/Active/Programming/03 NPU/03 Large Model/02-MemSAM/utils/evaluation.py", line 12, in <module>
      from medpy.metric.binary import hd as medpy_hd
  ModuleNotFoundError: No module named 'medpy'
  

  Solution: run $ pip3 install medpy

• the ./dataset/SAMUS/CAMUS_full/full.json is missing

  The CAMUS_full is not mentioned in the README.md file and the mentioned CAMUS dataset don't have class.json

  Error log

  (.venv) jason@Jasons-MacBook-Pro 02-MemSAM % python test_video.py
  task CAMUS_Video_Full checkpoints: checkpoints/CAMUS_full/your_checkpoint.pth
  Traceback (most recent call last):
    File "/Users/jason/01 Learn/Active/Programming/03 NPU/03 Large Model/02-MemSAM/test_video.py", line 76, in <module>
      test_dataset = EchoVideoDataset(opt.data_path, opt.test_split,  tf_val, img_size=args.encoder_input_size, frame_length=10,disable_point_prompt=args.disable_point_prompt, point_numbers=args.point_numbers)  # return image, mask, and filename
    File "/Users/jason/01 Learn/Active/Programming/03 NPU/03 Large Model/02-MemSAM/utils/data_us.py", line 810, in __init__
      with open(self.class_dict_file, 'r') as load_f:
  FileNotFoundError: [Errno 2] No such file or directory: './dataset/SAMUS/CAMUS_full/class.json'
  

What I've tried

1. Follow the README.md usage section
   1. download the CAMUS dataset and extract it to the ./dataset folder. E.g ./dataset/CAMUS_public
   2. download the EchoNet dataset and extract it to the ./dataset folder. E.g. ./dataset/EchoNet-Dynamic
   3. preprocess the CAMUS dataset and extract it to the ./temp_dir folder. E.g. ./temp_dir/CAMUS_public_256
   4. preprocess the EchoNet dataset and extract it to the ./temp_dir folder. E.g. ./temp_dir/EcoNet

[AGreenApple2002]

臣附议

[judyinf]

You can check the answers in #6