1 # Summary of commands - uncomment one and run this script
2 #### For the datasets in sections marked as "Automatic Download", dataset will be downloaded automatically downloaded before training begins. For "Manual Download", it is expected that it is manually downloaded and kept in the folder specified agaianst the --data_path option.
4 ## =====================================================================================
5 ## Training
6 ## =====================================================================================
7 #### Cityscapes Semantic Segmentation (Manual Download) - Training with MobileNetV2+DeeplabV3Lite
8 #python ./scripts/train_segmentation_main.py --dataset_name cityscapes_segmentation --model_name deeplabv3lite_mobilenetv2_tv --data_path ./data/datasets/cityscapes/data --img_resize 384 768 --output_size 1024 2048 --gpus 0 1
10 #### Cityscapes Semantic Segmentation (Manual Download) - Training with MobileNetV2+DeeplabV3Lite, Higher Resolution
11 #python ./scripts/train_segmentation_main.py --dataset_name cityscapes_segmentation --model_name deeplabv3lite_mobilenetv2_tv --data_path ./data/datasets/cityscapes/data --img_resize 768 1536 --rand_crop 512 1024 --output_size 1024 2048 --gpus 0 1
13 #### Cityscapes Semantic Segmentation (Manual Download) - Training with ResNet50+FPN
14 #python ./scripts/train_segmentation_main.py --dataset_name cityscapes_segmentation --model_name fpn_pixel2pixel_aspp_resnet50 --data_path ./data/datasets/cityscapes/data --img_resize 768 1536 --rand_crop 512 1024 --output_size 1024 2048 --gpus 0 1 \
15 #--pretrained https://download.pytorch.org/models/resnet50-19c8e357.pth
17 #### VOC Segmentation (Manual Download) - Training with MobileNetV2+DeeplabV3Lite
18 #python ./scripts/train_segmentation_main.py --dataset_name voc_segmentation --model_name deeplabv3lite_mobilenetv2_tv --data_path ./data/datasets/voc --img_resize 512 512 --output_size 512 512 --gpus 0 1
21 ## =====================================================================================
22 ## Validation
23 ## =====================================================================================
24 #### Validation - Cityscapes Semantic Segmentation (Manual Download) - Validation with MobileNetV2+DeeplabV3Lite - populate the pretrained filename in ??
25 #python ./scripts/train_segmentation_main.py --evaluate True --dataset_name cityscapes_segmentation --model_name deeplabv3lite_mobilenetv2_tv --data_path ./data/datasets/cityscapes/data --img_resize 384 768 --output_size 1024 2048 --gpus 0 1 \
26 #--pretrained ??
28 #### Inference - Cityscapes Semantic Segmentation (Manual Download) - Inference with MobileNetV2+DeeplabV3Lite - populate the pretrained filename in ??
29 #python ./scripts/infer_segmentation_main.py --dataset_name cityscapes_segmentation_measure --model_name deeplabv3lite_mobilenetv2_tv --data_path ./data/datasets/cityscapes/data --img_resize 384 768 --output_size 1024 2048 --gpus 0 1 \
30 #--pretrained ./data/modelzoo/semantic_segmentation/cityscapes/deeplabv3lite-mobilenetv2/cityscapes_segmentation_deeplabv3lite-mobilenetv2_2019-06-26-08-59-32.pth
32 #### Validation - VOC Segmentation (Manual Download) - Validation with MobileNetV2+DeeplabV3Lite - populate the pretrained filename in ??
33 #python ./scripts/train_segmentation_main.py --evaluate True --dataset_name voc_segmentation_measure --model_name deeplabv3lite_mobilenetv2_tv --data_path ./data/datasets/voc --img_resize 512 512 --output_size 512 512 --gpus 0 1
34 #--evaluate True --pretrained ???