search

samyak0210 / ViNet

ViNet Pushing the limits of Visual Modality for Audio Visual Saliency Prediction
MIT License
60 stars 19 forks source link

Change to support computation on CPU. #16

Open JulyFaraway opened 2 years ago

JulyFaraway commented 2 years ago

In model.load_state_dict, the "map_location" argument must also be set to match the device. Without this, the pretrained model calculated on the GPU cannot be used for calculations on the CPU.