andrespmd / gcn_classification Goto Github PK

Is it possible to provide submodule code instead of npy results? Like fasterrcnn and OCR's code after you edited which could let me get understand the process....

Hello, the link to the beverage dataset is not available

File "/data01/home/zpf/PycharmProjects/GCN_classification-master/src/data/context.py", line 28, in Context_dataset
with open(data_path+'/Context/data/split_'+ str(args.split) +'.json','r') as fp:
FileNotFoundError: [Errno 2] No such file or directory: '/data01/home/zpf/DataseContext/Context/data/split_1.json'

I have downloaded all the files and put them in the correct directory，but where is the split_1.json file? Thank you!

Thanks for you contribution! It is amazing. When i run test.py there is a problem.
ValueError: cannot reshape array of size 318242784 into shape (18488,36,2048)
and my command is python test.py bottles --model fullGCN_bboxes --ocr google_ocr --embedding fasttext --max_textual 15 --max_visual 36 --projection_layer mean --fusion concat

@AndresPMD Hello! I want to download your datasets, but I found all the google drive link is not available. Could you please update these links?

I want to use own data set.And how is the npy file of visual features generated?

Thank you for your codes, it is a good work! When I run the performance.py, a error "images_with_text.json" happen. But I find that this file not in the goolge share and not find the Code that Generate this file. How to get or Generate this file ?
Thank you very much!