GANs for Tabular Synthetic Data Generation (7.5)

แชร์
ฝัง
  • เผยแพร่เมื่อ 21 ก.ค. 2024
  • Typically GANs are used to generate images. However, we can also generate tabular data from a GAN. In this part, we will use the Python tabgan utility to create fake data from tabular data. Specifically, we will use the Auto MPG dataset to train a GAN to generate fake cars.
    Code for This Video:
    github.com/jeffheaton/t81_558...
    Course Homepage: sites.wustl.edu/jeffheaton/t8...
    Follow Me/Subscribe:
    / heatonresearch
    github.com/jeffheaton
    / jeffheaton
    Support Me on Patreon: / jeffheaton
  • วิทยาศาสตร์และเทคโนโลยี

ความคิดเห็น • 21

  • @firasnacef001
    @firasnacef001 ปีที่แล้ว +8

    I have been looking for data augmentation tools for my categorical data and the only ressources I found were for image data. Thank you so much !

  • @nirosetam
    @nirosetam ปีที่แล้ว +1

    Thanks Dr. Heaton, the code give me a direction for research work.

  • @flabbergastedfrog5851
    @flabbergastedfrog5851 5 หลายเดือนก่อน +4

    Hi
    This video was very helpful for me. But, there is a small correction i guess. In the "Training a GAN for Auto MPG" part the code has gan_params for parameters. But, as per tabgan, its actually gen_params. Please look into it
    Thank you

  • @sanyuktasharma5404
    @sanyuktasharma5404 3 หลายเดือนก่อน

    Thanks a lot for making this video in the "Training a GAN for Auto MPG" part the code has gan_params for parameters. But, as per tabgan, its actually gen_params. Just putting it here since I got stuck and read a comment and had my answer. But thankyou sir for making it!

  • @miladrahimi3283
    @miladrahimi3283 ปีที่แล้ว

    thanks dr. Heaton , this code is useful for me

  • @ShadArfMohammed
    @ShadArfMohammed 2 ปีที่แล้ว +1

    SMOTE and ADASYAN perform poorly on high dimentionality data (say 3000 features), can tabGAN outperform those two ?

  • @henkhbit5748
    @henkhbit5748 2 ปีที่แล้ว

    Interestting and useful. How's to generate master detail relationship?

  • @chyldstudios
    @chyldstudios 2 ปีที่แล้ว +1

    How do GANs compare to SMOTE for synthetic data generation?

  • @jonatapaulino
    @jonatapaulino ปีที่แล้ว

    Hey, congrats on the video. I want to generate synthetic data of emotions such as joy, treistness etc... Would it be possible to generate with GANS? I want to store this data to do some analysis. Thanks.

  • @cynthiac2174
    @cynthiac2174 2 ปีที่แล้ว +1

    Is it possible to do the same with WGAN?

  • @Cecbr93837
    @Cecbr93837 2 ปีที่แล้ว

    Thank you, really helpful !

  • @gustavojuantorena
    @gustavojuantorena 2 ปีที่แล้ว

    Helpful! Thank you

  • @kanjibrian1170
    @kanjibrian1170 2 ปีที่แล้ว +3

    can GANs generate tabular data from text?

  • @user-zj1eu9dm7r
    @user-zj1eu9dm7r 7 หลายเดือนก่อน

    Useful video !!

  • @Mohamm-ed
    @Mohamm-ed 2 ปีที่แล้ว +2

    Is ok to apply this method to EEG as it time series data ?

  • @jimimased1894
    @jimimased1894 2 ปีที่แล้ว +2

    tabgan utility super handy Jeff, thanks mate

  • @sarfkaff9770
    @sarfkaff9770 5 หลายเดือนก่อน

    Great

  • @AmitabhSuman
    @AmitabhSuman ปีที่แล้ว

    Hi Jeff.
    Great Video. Something similar, i have been looking for. But I need to get in touch with you for your guidance over a problem I am trying to solve. I took up AI and ML field just so that I can solve a problem at hand.
    Can you please help me with your mail ID or any means to contact you? It would be highly appreciated.
    Thanks.

  • @debasmitapal4084
    @debasmitapal4084 9 หลายเดือนก่อน

    from tabgan.sampler import GANGenerator
    This statement is not working. Getting error: 'ModuleNotFoundError: No module named 'ForestDiffusion'