is there any way to obtain the running time of the submited docker

Dear organizers,

It seems the submitted dockers very easy run out of time, which is the main reason of the error "The algorithm failed on one or more cases".

Althouth the docker works on the validation set, it could still run into error on the hidden testing set because the testing set includes some larger cases.

1. Is there any way to obtain the running time for each case of the submited docker? This infomation is useful for us to select the final model.
2. How to avoid the above situation (e.g., the submitted docker passed the validation phase but run into error on hidden large testing cases because of running time limitation)?

Hi, unfortunately there is no such way, but I can tell you the run time and how large the volume of the patient was. To solve this problem, I suggest resampling or cropping the axial slices to approx 600 if necessary.

Would it be possible to run the docker offline for these large testing cases?

The model has been trained on a specify a resolution. So we cannot resame the image to other spacing.

How to crop the axial slices to approx 600 (avoid moving the tumor part)?

Any suggested code is highly appreciated.

Hi, simply "throw" away the last few slices. As there are only a handfull of those large volumes this wont have any impact on the performance

Got it, thank you so much.