search

roahmlab / sel_map

Semantic ELevation (SEL) map is a semantic Bayesian inferencing framework for real-time elevation mapping and terrain property estimation from RGB-D images.
https://roahmlab.github.io/sel_map/
MIT License
107 stars 12 forks source link

Downloading pre trained network models from CSAIL link is prohibited #9

Open CookieAnt opened 9 months ago

CookieAnt commented 9 months ago

image Hello, I am unable to download the pre trained network model. Do you have any other address or cloud drive that you can share? thanks!

crankler commented 6 months ago

Have you access these model now? thanks for your reply.

BuildingAtom commented 6 months ago

Hi, apologies for the delayed response. We don't host them for redistribution ourselves, but I did a bit of digging and found that you can still download from CSAIL directly.

The general form is to use the base link

http://sceneparsing.csail.mit.edu/model/pytorch/

append the model name as another folder path, e.g. ade20k-resnet50dilated-ppm_deepsup/, then append encoder_epoch_20.pth and decoder_epoch_20.pth for the respective model weights.

For the example in the README, those two links would look like:

http://sceneparsing.csail.mit.edu/model/pytorch/ade20k-resnet50dilated-ppm_deepsup/encoder_epoch_20.pth
http://sceneparsing.csail.mit.edu/model/pytorch/ade20k-resnet50dilated-ppm_deepsup/decoder_epoch_20.pth

I'll leave this issue open until I update the repo so others can find this information. Thanks for asking!

crankler commented 5 months ago

Thank you for your share here