I want to train DRNET on the cityscapes dataset, here are the configurations for training:
DEVICE : cpu # device used for training and evaluation (cpu, cuda, cuda0, cuda1, ...)
SAVE_DIR : '/home/banana/Docs/VScode/Python/RSM_projects/Auto_Pilot/Kromka_Semantic_test/DDR-NET_output' # output folder name used for saving the model, logs and inference results
MODEL:
NAME : DDRNet # name of the model you are using
BACKBONE : DDRNet-23slim # model variant
PRETRAINED : '/home/banana/Docs/VScode/Python/RSM_projects/Auto_Pilot/Kromka_Semantic_test/pretrained_models/DDRNet-23slim.pth' # backbone model's weight
DATASET:
NAME : CityScapes # dataset name to be trained with (camvid, cityscapes, ade20k)
ROOT : '/media/banana/Linux_disk/Docs/data/cityscapes' # dataset root path
IGNORE_LABEL : 255
TRAIN:
IMAGE_SIZE : [1024, 1024] # training image size in (h, w)
BATCH_SIZE : 8 # batch size used to train
EPOCHS : 1 # number of epochs to train
EVAL_INTERVAL : 20 # evaluation interval during training
AMP : false # use AMP in training
DDP : false # use DDP training
LOSS:
NAME : OhemCrossEntropy # loss function name (ohemce, ce, dice)
CLS_WEIGHTS : false # use class weights in loss calculation
OPTIMIZER:
NAME : adamw # optimizer name
LR : 0.001 # initial learning rate used in optimizer
WEIGHT_DECAY : 0.01 # decay rate used in optimizer
SCHEDULER:
NAME : warmuppolylr # scheduler name
POWER : 0.9 # scheduler power
WARMUP : 10 # warmup epochs used in scheduler
WARMUP_RATIO : 0.1 # warmup ratio
EVAL:
MODEL_PATH : '/home/banana/Docs/VScode/Python/RSM_projects/Auto_Pilot/Kromka_Semantic_test/DDR-NET_output/ddrnet_23slim_city.pth' # trained model file path
IMAGE_SIZE : [1024, 1024] # evaluation image size in (h, w)
MSF:
ENABLE : false # multi-scale and flip evaluation
FLIP : true # use flip in evaluation
SCALES : [0.5, 0.75, 1.0, 1.25, 1.5, 1.75] # scales used in MSF evaluation
TEST:
MODEL_PATH : '/home/banana/Docs/VScode/Python/RSM_projects/Auto_Pilot/Kromka_Semantic_test/DDR-NET_output/ddrnet_23slim_city.pth' # trained model file path
FILE : 'assests/cityscapes' # filename or foldername
IMAGE_SIZE : [1024, 1024] # inference image size in (h, w)
OVERLAY : true # save the overlay result (image_alpha+label_alpha)
but when I start training I get the following error:
Traceback (most recent call last):
File "/home/banana/Docs/VScode/Python/RSM_projects/Auto_Pilot/Kromka_Semantic_test/DDR-NET/semantic-segmentation/tools/train.py", line 128, in <module>
main(cfg, gpu, save_dir)
File "/home/banana/Docs/VScode/Python/RSM_projects/Auto_Pilot/Kromka_Semantic_test/DDR-NET/semantic-segmentation/tools/train.py", line 42, in main
model.init_pretrained(model_cfg['PRETRAINED'])
File "/home/banana/Docs/VScode/Python/RSM_projects/Auto_Pilot/Kromka_Semantic_test/DDR-NET/semantic-segmentation/semseg/models/ddrnet.py", line 191, in init_pretrained
self.load_state_dict(torch.load(pretrained, map_location='cpu')['model'], strict=False)
KeyError: 'model'
I want to train DRNET on the cityscapes dataset, here are the configurations for training:
but when I start training I get the following error: