Emanuele Bugliarello comments

Results 5 comments of


                                            Emanuele Bugliarello

Is it possible to share the F1 score or accuracy of pre-training the models in their control setting in the image sentence alignment objective?

Hi, Can you clarify what is your goal? Also, you can run any analysis on _all_ the models pretrained in the control setup (cf. [here](https://github.com/e-bug/volta/blob/main/MODELS.md#checkpoints-by-random-seed))

On the visual token added to linguistic tokens in VLBertEmbeddings class

Hi Iki-san, Thanks for pointing this out, you are right. I don't think this impacts performance much but I'll try to fix it. I cannot however just fix the code...

Can i use this script to create Resnet 56 and 110 models

I have implemented them in my fork at: https://github.com/e-bug/pytorch-cifar

where are the Chinese VQA Text Data

Hi, the above dataset only contains 239,941 question-answer pairs, which are only for the `train` and `val` splits. 1. The results in the paper are for 250,569 pairs. 1a. Are...

NLVR2 images

Thanks a lot for the prompt fix!