search

vandal-vpr / vg-transformers

Official Repository of "Learning Sequential Descriptors for Sequence-based Visual Place Recognition "
MIT License
39 stars 5 forks source link

Timesformer #12

Closed Steven-jiaqi closed 1 year ago

Steven-jiaqi commented 1 year ago

With your help, I have successfully run the cct+seqvlad experiment. But when I was running the Timesformer experiment, I found some errors: AttributeError: ‘Namespace' object has no attribute 'features_dim' It happens in "vg-transformers/tgv/models/tgv_net.py" line 43 self.meta = {'outputdim': args.features_dim} And i can not find the 'args.features_dim' variant in the code project. Should I assign a value to this parameter in the parser.py? Thanks for your reply!

ga1i13o commented 1 year ago

Hello, I was not able to reproduce your issue. The args.features_dim value is assigned in the function below:

https://github.com/vandal-vpr/vg-transformers/blob/25537ed1689d4b5401113f6c4476fb8f0cf28158/tvg/models/tvg_net.py#L159

Specifically at this line for the TimeSformer https://github.com/vandal-vpr/vg-transformers/blob/25537ed1689d4b5401113f6c4476fb8f0cf28158/tvg/models/tvg_net.py#L179

Have you pulled the last version of the code?

ga1i13o commented 1 year ago

Solved in https://github.com/vandal-vpr/vg-transformers/issues/13