search

duyhominhnguyen / LVM-Med

[NeurIPS 2023] Release LMV-Med pre-trained models
188 stars 24 forks source link

python train_segmentation.py -c ./dataloader/yaml_data/buid_endtoend_R50.yml has a error #19

Closed buzhidaozenmeqimingzi closed 6 months ago

buzhidaozenmeqimingzi commented 7 months ago

Error shows below: INFO: Using device cuda:5

Using pre-trained models from lvm-med-resnet Traceback (most recent call last): File "train_segmentation.py", line 14, in train_2d_R50(yml_args, cfg) File "/root/autodl-tmp/LVM-Med/segmentation_2d/train_R50_seg_adam_optimizer_2d.py", line 224, in train_2d_R50 net = smp.Unet(encoder_name="resnet50", encoder_weights=cfg.base.original_checkpoint, File "/root/miniconda3/envs/lvm_med/lib/python3.8/site-packages/segmentation_models_pytorch/unet/model.py", line 59, in init self.encoder = get_encoder( File "/root/miniconda3/envs/lvm_med/lib/python3.8/site-packages/segmentation_models_pytorch/encoders/init.py", line 38, in get_encoder settings = encoders[name]["pretrained_settings"][weights] KeyError: 'lvm-med-resnet

buzhidaozenmeqimingzi commented 7 months ago

Hello!I had come across a problem as above.I don't know whether it related to source code or not.Please inform me why it happened, I'll be appreciated.