search

salaniz / pytorch-gve-lrcn

PyTorch implementations for "Generating Visual Explanations" (GVE) and "Long-term Recurrent Convolutional Networks" (LRCN)
MIT License
92 stars 22 forks source link

gve trained model #18

Open hannahgym opened 3 years ago

hannahgym commented 3 years ago

can i have access to the trained gve model?

i know that caffemodel version is available here: https://drive.google.com/drive/folders/1nU9ATTLHAM6_jz-K6hoVlDzNrFcOXtyH

Ellyuca commented 2 years ago

Hi @Hengameh400. Did you manage to find the trained gve pytorch model? Thanks.

salaniz commented 1 year ago

Since this is a reimplementation and nothing official, I don't see an easy way of providing weights that would reflect the original caffe model.

There is some resources that can convert weights (e.g. deep-learning-model-convertor, MMdnn) but the implementation could still be slightly different and not perfectly match each layer. This implementation was done based on the paper and not a 1-to-1 translation from the caffe code.

If you find a way to convert them and open a pull request, I would be happy to include the code in this repository.

I got similar, but not the exact same results as in the paper, so I wouldn't want to upload my pre-trained models.