Pretrained Encoder link

[BrianPugh]

In the README, the pretrained encoder cmd should be:

wget https://storage.googleapis.com/vit_models/imagenet21k/R50%2BViT-B_16.npz

not the current:

wget https://storage.googleapis.com/vit_models/imagenet21k/R50-ViT-B_16.npz

[BrianPugh]

while we're at it, the mkdir command should be:

mkdir -p ./model/vit_checkpoint/imagenet21k 

[BrianPugh]

finally, the last command should be:

mv R50+ViT-B_16.npz model/vit_checkpoint/imagenet21k/

confirmed by the config you are using here

[BrianPugh]

and the prepare nyu commands can be shorted to:

mkdir -p pytorch/dataset/nyu_depth_v2
python utils/download_from_gdrive.py 1AysroWpfISmm-yRFGBgFTrLy6FjQwvwP pytorch/dataset/nyu_depth_v2/sync.zip
cd pytorch/dataset/nyu_depth_v2
unzip sync.zip

[ygjwd12345]

In the README, the pretrained encoder cmd should be:

wget https://storage.googleapis.com/vit_models/imagenet21k/R50%2BViT-B_16.npz

not the current:

wget https://storage.googleapis.com/vit_models/imagenet21k/R50-ViT-B_16.npz

it is according to ViT

imagenet21k pre-train

wget https://storage.googleapis.com/vit_models/imagenet21k/{MODEL_NAME}.npz

[ygjwd12345]

and the prepare nyu commands can be shorted to:

mkdir -p pytorch/dataset/nyu_depth_v2
python utils/download_from_gdrive.py 1AysroWpfISmm-yRFGBgFTrLy6FjQwvwP pytorch/dataset/nyu_depth_v2/sync.zip
cd pytorch/dataset/nyu_depth_v2
unzip sync.zip

Thanks

[BrianPugh]

the MODEL_NAME has a +, not a - after R50

[ygjwd12345]

Sorry it's my mistake

[ygjwd12345]

the MODEL_NAME has a +, not a - after R50

It has be fixed.