when I was testing using
medperf test run --demo_dataset_url synapse:syn52276402 --demo_dataset_hash "16526543134396b0c8fd0f0428be7c96f2142a66" -p ./prep_segmentation -m "/home/hln0895/medperf/menunet/mlcube/" -e ./eval_segmentation --offline --no-cache
================================================================================================================
I got errors as shown as follows.
Traceback (most recent call last):
File "/mlcube_project/mlcube.py", line 29, in
app()
File "/mlcube_project/mlcube.py", line 19, in infer
raise NotImplementedError("The infer method is not yet implemented")
NotImplementedError: The infer method is not yet implemented
ERROR:
message: DockerRun runner failed to run MLCube.
description: Error occurred while running MLCube task (docker=docker, run_args= --network=none, env_args=, volumes=--volume /home/hln0895/.medperf/api_medperf_org/data/72d37eb6bd4ecd17cb3ecc04c57c4fefa6b02e26/data:/mlcube_io0 --volume /home/hln0895/.medperf/api_medperf_org/cubes/26119e961541e940da73bd2fd045f3102d9cd985/workspace:/mlcube_io1 --volume /home/hln0895/.medperf/api_medperf_org/cubes/26119e961541e940da73bd2fd045f3102d9cd985/workspace/additional_files:/mlcube_io2 --volume /home/hln0895/.medperf/api_medperf_org/cubes/26119e961541e940da73bd2fd045f3102d9cd985/workspace/additional_files/model_folder:/mlcube_io3 --volume /home/hln0895/.medperf/api_medperf_org/tmp68843/1443302904:/mlcube_io4, image=menunet, task_args=['infer', '--data_path=/mlcube_io0', '--parameters_file=/mlcube_io1/parameters.yaml', '--weights=/mlcube_io2/weights.pth', '--model_folder=/mlcube_io3', '--output_path=/mlcube_io4']).
context: {'status': 'exited', 'code': 1, 'cmd': 'docker run --network=none --volume /home/hln0895/.medperf/api_medperf_org/data/72d37eb6bd4ecd17cb3ecc04c57c4fefa6b02e26/data:/mlcube_io0 --volume /home/hln0895/.medperf/api_medperf_org/cubes/26119e961541e940da73bd2fd045f3102d9cd985/workspace:/mlcube_io1 --volume /home/hln0895/.medperf/api_medperf_org/cubes/26119e961541e940da73bd2fd045f3102d9cd985/workspace/additional_files:/mlcube_io2 --volume /home/hln0895/.medperf/api_medperf_org/cubes/26119e961541e940da73bd2fd045f3102d9cd985/workspace/additional_files/model_folder:/mlcube_io3 --volume /home/hln0895/.medperf/api_medperf_org/tmp68843/1443302904:/mlcube_io4 menunet infer --data_path=/mlcube_io0 --parameters_file=/mlcube_io1/parameters.yaml --weights=/mlcube_io2/weights.pth --model_folder=/mlcube_io3 --output_path=/mlcube_io4'}
? Model MLCube failed.
WARNING: Failed to premanently cleanup some files. Consider deleting '/home/hln0895/.medperf/.trash/68843' manually to avoid unnecessary storage.
Created by hui lin huilin @aneeshashmi Do you mean Predictions number don't match labels? I am also getting the same error.
Did u manage to solve it? @huilin File "/mlcube_project/metrics.py", line 28, in check_for_segmentation
raise ValueError("Predictions number don't match labels")
ValueError: Predictions number don't match labels
In the path, the nii.gz files demo/labels/ do not just have three values, like 0,1,2,3
Before testing, read https://docs.medperf.org/mlcubes/mlcube_models/, there are examples below.I'm guessing you didn't change the code for mlcube.py