snsd0805/adversarial_VLNBERT
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

update readme

Browse Source
This commit is contained in:
Yicong Hong 2021-01-14 16:27:21 +11:00
parent 2588537505
commit d95d8943f0
1 changed files with 9 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
README.md
View File

@ -29,6 +29,15 @@ Please follow the instructions below to prepare the data in directories:
- MP3D image features: `img_features`
- Download the [Scene features [4.2GB]](https://www.dropbox.com/s/85tpa6tc3enl5ud/ResNet-152-places365.zip?dl=1) (ResNet-152-Places365).
### Initial OSCAR and PREVALENT weights
Please refer to [vlnbert_init.py](https://github.com/YicongHong/Recurrent-VLN-BERT/blob/main/r2r_src/vlnbert/vlnbert_init.py) to set up the directories.
- Pre-trained [OSCAR](https://github.com/microsoft/Oscar) weights
- Download the `base-no-labels` following [this guide](https://github.com/microsoft/Oscar/blob/master/DOWNLOAD.md).
- Pre-trained [PREVALENT](https://github.com/weituo12321/PREVALENT) weights
- Download the `pytorch_model.bin` from [here](https://drive.google.com/drive/folders/1sW2xVaSaciZiQ7ViKzm_KbrLD_XvOq5y).
### Trained Network Weights
- Recurrent-VLN-BERT: `snap`