Sagar Vinodababu
Sagar Vinodababu
Hi @Lzhushuai, sorry for the really delayed reply. Yes, what @soloist97 said is right. I do, however, remove it in the final evaluation in [`eval.py`](https://github.com/sgrvinod/a-PyTorch-Tutorial-to-Image-Captioning/blob/master/eval.py) so it doesn't contribute to...
Hi, thanks! Sorry for the late reply, I've been traveling. My understanding of hard attention and VAEs is _very_ superficial, sadly. Thanks for the question - I will now finally...
Hi @dwang68, To add to what @kmario23 said, the reason they are in `[0, 255]` is because true color images have 256 values (8-bit) for each channel. It's common to...
Hey, sorry, I've been really busy and I'll respond to your email (and here) in the next day or two. I don't remember the loss values, especially the training loss...
What's the error you get (with traceback)?
Hello, it's been a while since I read the paper, so I don't remember my thought process regarding why I went ahead with a scale of 0,1 for `conv4_3`. Looking...
> You need to extract the zip first (I think) but I'm having issues with an archive header when I try to do so. Hi, you do not need to...
Hi @NicolasHarding, I just looked at it, and I think you are right. Thanks for pointing this out. I was careless and the priors won't be clipped. Therefore, some of...
@AndreiMoraru123 thanks for confirming! I'll check out HuggingFace, this idea never occurred to me, thank you.
> @sgrvinod I sincerely apologize for bothering you multiple times, but it seems that this is not correct : > In ```class MultiBoxLoss``` : > > ```python > self.smooth_l1 =...