main.py: error: unrecognized arguments: --fold=0 about research-contributions HOT 5 OPEN

In addition, I train the model myself and get the weight model Pt, then use the following command:
python test. py --infer_ overlap=0.5 --data_ dir=/data1/dataset/wdDATASET/BTCV/abdomen/ --pretrained_ dir='./ pretrained_ models/' --saved_ checkpoint=ckpt --pretrained_ model_ name=model. pt
The result produced an error:
Unexpected key(s) in state_ dict: "epoch", "best_acc", "state_dict".
I will --saved_ checkpoint=torchscript
The result produced an error:
PytorchStreamReader failed locating file constants. pkl: file not found
Excuse me, how to solve this? Thank you very much!!

from research-contributions.

Hello,
I faced the same issue, and I simply think they did not provide either the version with the k-fold nor the code for ensemble evaluation. In addition, I believe ensemble evaluation is fundamental to achieve high performance.

from research-contributions.

Hi @BelieferQAQ , which repository did you try ?

from research-contributions.

Hi @BelieferQAQ , which repository did you try ?

UNETR,

Using the default values for hyper-parameters, the following command can be used to initiate training using PyTorch native AMP package:

python main.py
--feature_size=32
--batch_size=1
--logdir=unetr_test
--fold=0
--optim_lr=1e-4
--lrschedule=warmup_cosine
--infer_overlap=0.5
--save_checkpoint
--data_dir=/dataset/dataset0/

from research-contributions.

@ahatamiz Hi,
Thank you for your great work. Could you please help how I can do k-fold validation using the commands? I am facing an error saying --fold in not recognized. I am using a private dataset and not a pre-trained one.
Thanks.

from research-contributions.