search

ExplainableML / AVCA-GZSL

This repository contains the code for our CVPR 2022 paper on "Audio-visual Generalised Zero-shot Learning with Cross-modal Attention and Language"
MIT License
34 stars 1 forks source link

hardware & runtime #1

Closed Qichen98fly closed 2 years ago

Qichen98fly commented 2 years ago

So what/How many gpus did you use for training each dataset? How much time for each one

MerceaOtniel commented 2 years ago

We used 1 GPU to train all networks for all datasets. The training time was shorter than 1 hour per method per dataset [f-vaegan-d2 took longer]. We fit the model on 2080ti.

Qichen98fly commented 2 years ago

And how about the same questions for feature extraction>

MerceaOtniel commented 2 years ago

Feature extraction took at most 2 days.

Qichen98fly commented 2 years ago

Feature extraction took at most 2 days.

Thanks for letting me know