yuanzhe1999/OpenECAD_Project
1
0
Fork 0
mirror of https://github.com/Yuki-Kokomi/OpenECAD_Project.git synced 2026-02-04 00:33:22 -05:00
Code Issues Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
Yuki-Kokomi
2024-11-07 16:12:59 +08:00
committed by GitHub
parent 7d1be2448b
commit a65d661128
1 changed files with 6 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
README.md
View File

@@ -11,6 +11,8 @@ The dataset and model weights can be accessed at https://huggingface.co/collecti
To generate the OpenECAD dataset and train using other CAD datasets, please start from scratch (this project uses DeepCAD as an example). To generate the OpenECAD dataset and train using other CAD datasets, please start from scratch (this project uses DeepCAD as an example).
If you would like to directly try out our model's generation capabilities, you can run `aioscript.sh` and follow the step-by-step instructions provided.
## Step 1. Use the Bethany tool to convert the DeepCAD dataset into the OpenECAD dataset. ## Step 1. Use the Bethany tool to convert the DeepCAD dataset into the OpenECAD dataset.
First, download all the JSON files from the DeepCAD dataset and place them in the same directory. (Download link: http://www.cs.columbia.edu/cg/deepcad/data.tar) First, download all the JSON files from the DeepCAD dataset and place them in the same directory. (Download link: http://www.cs.columbia.edu/cg/deepcad/data.tar)
@@ -87,6 +89,10 @@ If the environment is not installed, please first follow the steps mentioned in
You can use the model with the WebUI provided by TinyLLaVA: You can use the model with the WebUI provided by TinyLLaVA:
```shell ```shell
# You can download our models using git-lfs.
# git lfs install
# git clone https://huggingface.co/Yuki-Kokomi/OpenECADv2-SigLIP-2.4B
python tinyllava/serve/app.py --model-path <Your_Model_Path> python tinyllava/serve/app.py --model-path <Your_Model_Path>
``` ```