Xianshun Chen
Xianshun Chen
@babyhuzi111 one possibility may be the tokenizer, if you can share with me your training chinese text file. i can try it with my models and let you know.
i think recursive-rnn-1 or recursive-rnn-2 should be good for the start
@uhauha2929 the text body and summarized text use different vocabulary, which might explain what you observed. Also the max_sequence_length is set on the text body, meaning it does not read...
@ersinyar i used fake_or_real_news.csv while training these models for the scripts in the demo folder
one way i think can try is to encode ur image into a fixed-length vector (that is feature extraction) using a pretrained model such as VGG or inception resnet, then...
do folder structures of your mp4 files must matched that of the UCF-101 avi files and also your directory names should be updated in the code instead of UCF-101, please...
@alnsn Thank you for the suggestion! i will try to add this in these two days.
Sorry pretty busy recently with my works, did not have a chance to look at how to resolve this. One way that you may try is to change the Pdf2Image.PrintQuality...
@davinchi7 thanks for the demo.pdf, i think you are right about the font. i was tuning the fonts for the pdf convert, still working on it. below are some of...
Hi @gursimar Many thanks for notifying me the issue and sharing with me your code, I will look into this and get back to you :)