miyyer / comics Goto Github PK

Hi,

In the dataset and in this repository, I could not see any file indicating which number corresponds to which comic series. Is there any information on the names of the comics that are stored in the folders 0 to 3958?

The following links seems do not be valid anymore:
https://obj.umiacs.umd.edu/comics/raw_panel_images.tar.gz
https://obj.umiacs.umd.edu/comics/vgg_features.h5

I was having some trouble understanding what all of the input variables are and was hoping explanations could be provided.

This is from text_cloze.py, with comments annotated by my understanding/questions

    # input theano vars
    # these are just the images from the context panels
    in_context_fc7 = T.tensor3(name='context_images') 

    # unsure of what the bbmask contains vs the context_bb 
    in_context_bb = T.tensor4(name='context_bb')
    in_bbmask = T.tensor3(name='bounding_box_mask')

    # is in_context the actual text from the context panels?
    in_context = T.itensor4(name='context')

    # what is this mask vs the bb mask?
    in_cmask = T.tensor4(name='context_mask')

    # are these the image and bb for the answer panel?
    in_answer_fc7 = T.matrix(name='answer_images')
    in_answer_bb = T.matrix(name='answer_bb')

    # I see that answers is of shape 3 x max_words, where 3 is the num of context panels, but what do the numbers in this tensor mean?
    # when I printed it out, it looked like
 # [[[ 5547    17  1547 ...     0     0     0]
 # [  776 20000 20000 ...     0     0     0]
 # [  102     4    13 ...     0     0     0]]
   in_answers = T.itensor3(name='answers')

    # what is the mask for?
    in_amask = T.tensor3(name='answer_mask')

    # the labels indicate which answers are the correct ones 
    in_labels = T.imatrix(name='labels')

Hello author and everyone.
Allows me make a question that difficult for me.
"to train models after preprocessing (example for text cloze):
python models/text_cloze.py (make sure to run on GPU; see run.sh for our theano flags)
see description of hyperparameters by running python models/text_cloze.py --help
note that low-quality data is only filtered out in dev/test data (by throwing out examples with too many UNK tokens).
during training, all data is used."
I followed above comments, I saw text_cloze.py in models folder. I typed "python text_cloze.py" and had error:

Please help me. I also know I need to setup run.sh but I am not similar with setup them.

Do I show what path in my computer? Let's me know?

Thank you very much.

the egg_features.h5 is fine-tuned? or just same with vgg features provided by eras or tf?

[to notify]

Could you please provide vocab file necessary for text data?
vdict, rvdict = pickle.load(open('data/comics_vocab.p', 'rb'))