NeuralBabyTalk
NeuralBabyTalk copied to clipboard
jiasenlu
Pytorch code of for our CVPR 2018 paper "Neural Baby Talk"
I can generate captions for my images. can you explain about that?
Excuse me, do you use the self-critical and cider optimizer about the paper result? If we use self-critical, we should use the training process first? And about the finetune cnn...
I can not seem to find the the code regarding the accuracy evaluation for the robust coco split. It would be great to see it in order to be able...
In my environment (pytorch 0.4.0), it will report error for line 78: word = itow[str(ix)] Because ix is a tensor, so it is better to use ix.item() to replace the...
eg:a [ backpack ] on a [ cheesecake ] on a street。 But the picture does not detect the two words， [ backpack ] ， [ cheesecake ]
Running a pre-trained model on COCO, I'm getting strange predicted captions (examples shown below). Has anyone else run into this issue? ```image 391895: wipe liner undergoing wipe mutt topless wipe...
Hello. I trained the model with the default settings (Standard Image Captioning), but I got a lower score than the reported score in you paper. Here is my result: Bleu_1:...
excuse me, I get this error when I train coco with pytorch==0.4 .1 python==3.5.2 , I change batch_size from 20 to 10.how can i solve the problem? thanks File "/home/ubuntu/MyFiles/zwr/NeuralBabyTalk/misc/utils.py",...
The `main.py` has pdb.set_trace() in the case of `opt.inference_only`, [here in code](https://github.com/jiasenlu/NeuralBabyTalk/blob/4a2ef428ec9f251a1eb898cc0c828a6ef1c55e69/main.py#L176) It is a debuger code for python programs. Because of this COCO evaluation command you give stops after...
