zzw1123

Thanks for your contribution! But I have a question that during training, the total number of iteration of every epoch is ：n_iters_per_epoch = int(np.ceil(float(n_examples)/self.batch_size)) since there are about 5 captions...

Thanks for your contribution! But when I use the the command "python train.py" to run the model after all preprocessing procedures are done, my program is stuck without any messages...

Questions about training ACGAN

1

Thanks for your contribution! A problem occurs when I am training ACGAN without tuning any parameters which is shown as follows: As the training process goes on, the loss of...

Questions in the procedure of validation on lfw

2

Thanks for your contribution. But I have a doubt that before feeding the testset into the model we trained already, should we first detect faces in the images of lfw?In...

Thanks for your contribution! A problem occurs when I am training ACGAN without tuning any parameters which is shown as follows: As the training process goes on, the loss of...

Error when unzipping MHP v2 dataset

3

There are not any errors while downloading the dataset, however, it is said that "the data is damaged" when I unzip it on Windows. (The error exists whether downloaded from...

How to get captions of the image?

1

Thanks for your contribution! A question confused me for a while, how do you get the captions of all the image? Because as far as I know the Oxford-102 data...

Hi, I notice that in Non_local module, the original matrix multiplication is replaced by concat and conv... I wonder why you do like this? Could you please explain it? Thanks!...

cannot identify image file <_io.BytesIO object at 0x7f2d80b2e770>

5

Thanks for your contribution! But when I run the code, an error occurs: Traceback (most recent call last): File "runtime.py", line 42, in trainer.train(args.cls) File "/home/zzw/program/text2img/text-to-Image-Synthesis-pytorch/trainer.py", line 65, in train...

Questions about detailed attention map during test phase

2

Hi, thanks for your contribution! In Section 4.2 in the paper, you mentioned that ``` we randomly select a channel of attention maps in each iteration in training stage, and...