test_v2.sh

######## Outdoor Model Configuration ########
    
# outdoor_path = 
# outdoor_model = 'deeplabv3plus_mobilenet' 
# lr = 0.01
# crop_size = 224
# batch_size = 1
# output_stride = 16
# outdoor_checkpoint = real/DeepLabV3Plus-Pytorch/checkpoints/best_deeplabv3plus_mobilenet_cityscapes_os16.pth
# outdoor_save_path =  

# ######## Indoor Model Configuration ########

# indoor_path = ./indoor/img_data
# indoor_model_cfg = indoor/config/ade20k-mobilenetv2dilated-c1_deepsup.yaml 
# indoor_model = ade20k-mobilenetv2dilated-c1_deepsup  
# outdoor_checkpoint = indoor/epoch_20.pth
# outdoor_save_path = ,
# test_mode = "outdoor"    ## indoor or outdoor ?

# if test_mode == "outdoor":
#     python3 real/DeepLabV3Plus-Pytorch/main.py --model $outdoor_model \
#     --crop_val --lr $lr --crop_size $crop_size --batch_size $batch_size \
#     --output_stride $output_stride --ckpt $outdoor_checkpoint \
#     --test_only --save_val_results

# elif test_mode == "indoor":
#     python3 -u indoor/test.py \
#     --imgs config["indoor_path"] \
#     --cfg config["indoor_model_cfg"] \
#     DIR config["indoor_model"] \
#     TEST.result config["outdoor_save_path"] \
#     TEST.checkpoint config["outdoor_checkpoint"]

python3 real/DeepLabV3Plus-Pytorch/main.py --model deeplabv3plus_mobilenet \
    --crop_val --lr 0.01 --crop_size 224 --batch_size 2 \
    --output_stride 16 --ckpt checkpoint_real/real_channel.pth \
    --test_only --save_val_results