search

Cadene / vqa.pytorch

Visual Question Answering in Pytorch
716 stars 177 forks source link

I do not have a demo server #30

Closed youngjaean closed 6 years ago

youngjaean commented 6 years ago

Warning: 565/930911 words are not in dictionary, thus set UNK Warning fusion.py: no visual embedding before fusion Warning fusion.py: no question embedding before fusion Warning fusion.py: no visual embedding before fusion Warning fusion.py: no question embedding before fusion Warning train.py: no optim checkpoint found at 'logs/vqa/mutan_noatt_train/best_optim.pth.tar' This is an error.

I didn't have enough hard disk capacity, so i deleted the raw data Does this matter?

Cadene commented 6 years ago

Warning: 565/930911 words are not in dictionary, thus set UNK

just for information

Warning fusion.py: no visual embedding before fusion Warning fusion.py: no question embedding before fusion Warning fusion.py: no visual embedding before fusion Warning fusion.py: no question embedding before fusion

just for information (the same layer is used for the last fusion and the attention, both are a bit different and we add a warning, but you could simply remove it)

Warning train.py: no optim checkpoint found at 'logs/vqa/mutan_noatt_train/best_optim.pth.tar'

It does matter only if you need to resume the training

Deleting the raw data is not a probem as long as you keep the raw directory (with no content). https://github.com/Cadene/vqa.pytorch/blob/master/vqa/datasets/utils.py#L15

youngjaean commented 6 years ago

I do not understand. You do not need raw data? Do you just keep the directory? Does that solve the error on the demo server? Thank you for the reply.