search

GeorgeSeif / Semantic-Segmentation-Suite

Semantic Segmentation Suite in TensorFlow. Implement, train, and test new Semantic Segmentation models easily!
2.51k stars 878 forks source link

how to run test.py? #214

Open vkes2541 opened 5 years ago

vkes2541 commented 5 years ago

hello i'm korean student.

yesterday i tried to run train.py and latest_model_DeepLabV3_CamVid.ckpt was made. and to try to run test.py, i tried this command "python test.py --models DeepLabV3 --checkpoint_path checkpoints/latest_model_DeepLabV3_CamVid.ckpt. but i got this error "subprocess.CalledProcessError: Command '['python', 'utils/get_pretrained_checkpoints.py', '--model=ResNet101']' returned non-zero exit statutus 1."

i saw the issues related to this error. as the answer to this error, see the FAQ in issues template. please let me know about this FAQ and how to run this code.

i'm sorry that my english level is low. please help me.

grbagwe commented 5 years ago

Run this (Also make sure that your crop size is correct): python test.py --models DeepLabV3 --checkpoint_path checkpoints/latest_model_DeepLabV3_CamVid/.ckpt

INJNainggolan commented 5 years ago

Hi,could you deal this question? My premodel is resnet_50, and the model is BiSeNet, When I run train.py, it operate normally, but when I run test.py and predict.py, it download premodel resnet_101,why?

liu-zg15 commented 4 years ago

Hi,could you deal this question? My premodel is resnet_50, and the model is BiSeNet, When I run train.py, it operate normally, but when I run test.py and predict.py, it download premodel resnet_101,why?

Please revise the default pre-trained model in model_builder.