kyfafyd / mirrorgan Goto Github PK

hello, I want to do some experiments on veri776 datasets. I have classified the pictures and the text I wrote into folders by category. But I don't have pickle file and json file, I would like to ask if I can generate these contents after I create these empty files and run preprocess.py? I would like to know what is caption.pickle used for? I opened the caption.pickle of birds and found that it has four parts, which represent train_captions, test_captions, ixtoword, wordtoix. I want to know what the numbers in train_captions or test_captions represent?
btw,i think you are Chinese , can I communicate with you in Chinese? it can express more clearly

image_encoder.pth,decoder.ckpt and encoder.ckpt are 404.
Can you upload it again?
Thanks

Could you please let me know what is the purpose of line
for c in filelist2:
if c == 'Muscular' or c == 'Circulatory' or c == 'Dental_Tooth':
continue
ff = os.listdir('data/health/CUB_200_2011/images/' + c)

The link to image_encoder.pth pf STEM has failed. Can you please upload it again?

rt.

Hello sir
Do you have bird/STREAM pretrained models? 【cnn/rnn_encoder.ckpt with torch.Size([5450, 512])】, I have some difficult in training the image caption model.
Thank you very much

I‘d like to train the model and when I try to run the GLAM main.py, the error occurs, I find it on the google,but i failed to solve the problem, can you teach me why this error occur and what can I do to solve it. Thanks!

I aim to use this model implementation as a black box and to input my own caption and should get than an image as an output.
Thanks

Hello sir
Do you have a cub / coco pretrained model?The author did not provide，I just want to use it for comparative experiments
【MirrorGAN pretrained model, not the stem / stream model】
Thanks

I'm trying to train with my own dataset, but I'm having some problems with preprocessing，Can you tell me What are the functions of list1.txt and list2.txt in preprocess.py？