vhad/SoftGroup
1
0
Fork 0
mirror of https://github.com/botastic/SoftGroup.git synced 2025-10-16 11:45:42 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
Thang Vu 2022-03-07 20:53:50 +09:00 committed by GitHub
parent 714bacd0c9
commit 6532747902
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
README.md
View File

@ -35,15 +35,15 @@ Download the pretrained HAIS model at [here](https://drive.google.com/file/d/1XG
### Training S3DIS dataset
First, finetune the pretrained HAIS point-wise prediction network (backbone) on S3DIS.
```
python train.py --config config/softgroup_backbone_s3dis.yaml
python train.py --config config/softgroup_fold5_backbone_s3dis.yaml
```
Then, train model from frozen backbone.
```
python train.py --config config/softgroup_default_s3dis.yaml
python train.py --config config/softgroup_fold5_default_s3dis.yaml
```
### Training ScanNet V2 dataset
Training on ScanNet doesnot require finetuning the backbone. Just frozen pretrained backbone and train the model.
Training on ScanNet doesnot require finetuning the backbone. Just freeze pretrained backbone and train the model.
```
python train.py --config config/softgroup_default_scannet.yaml
```