search

SuperMedIntel / Medical-SAM-Adapter

Adapting Segment Anything Model for Medical Image Segmentation
GNU General Public License v3.0
1.02k stars 90 forks source link

Training stuck after loading 3d data #20

Open CarinaL99 opened 1 year ago

CarinaL99 commented 1 year ago

Training stuck after loading 3d data Snipaste_2023-06-08_16-43-17

python train.py -net sam -mod sam_adpt -exp_name msa-3d-sam-btcv -sam_ckpt ./sam_vit_b_01ec64.pth -image_size 1024 -b 1 -dataset decathlon -thd True -chunk 1 -data_path /mnt/data/abdomen -num_sample 1

i run the program, but it gets stuck after loading the 3d data. How to solve it

AmrinKareem commented 1 year ago

Hi @CarinaL99 @WuJunde Does the 3D data have an associated dataset class? I couldn't find it in dataset.py.

CarinaL99 commented 1 year ago

Hi @CarinaL99 @WuJunde Does the 3D data have an associated dataset class? I couldn't find it in dataset.py.

I just use BTCV dataset following the training 3d tutorial as README.md

AmrinKareem commented 1 year ago

Hi @CarinaL99 Did you solve the problem? I am facing the same problem too. My training is stuck after loading the 3D data. @WuJunde Can you please advise? Please let me know. Thanks!

visionbike commented 2 months ago

I met this problem, chunk size should be > 1, I used 3 and it work well