Pretrained weights for pytorch version

dougsm / ggcnn

Generative Grasping CNN from "Closing the Loop for Robotic Grasping: A Real-time, Generative Grasp Synthesis Approach" (RSS 2018)

BSD 3-Clause "New" or "Revised" License

484 stars 138 forks source link

Pretrained weights for pytorch version #27

Closed ancorasir closed 4 years ago

ancorasir commented 4 years ago

Thanks the sharing the code with the community. Would you share the pretrained weights for pytorch as well? I'd like to test trained model on images of novel object for example.

dougsm commented 4 years ago

Hi @ancorasir Sorry for the super slow reply, I completely missed this.
Pretrained weights for some different conditions are available here: https://github.com/dougsm/ggcnn/releases/download/v1.0/ggcnn2_weights.zip

Note that if you have a specific application in mind, retraining for that scenario might provide better results, especiall aspects like field of view and data augmentation.

trungpham2606 commented 4 years ago

@dougsm Seem there's something wrong that I cant load your saved file. Can you specify the pytorch version you used for this project ? Your saved model cant be loaded in other PCs excluding yours.