Inspiration
pytorch have a great role in making GAN module. TESTGAN pytorch is a pytorch framework for GENERATIVE ADVERSARIAL NETWORK GAN* Textgan serve as a benchmarking platform to support research on GAN based text generation models. Since most GEN based text generation models are implemented by tensorflow, TESTGAN can help these who get use to pytorch to enter text generation field faster. Do it Inspired us to use great pytorch with other libraries such as num,ntlk to make GAN* MODULE with pytorch
What it does
Textgan pytorch help those who get used to pytorch to enter the text generation field faster as those who not use tensorflow.
How we built it
It's build with such libraries as following... .Python >=1.1.0 .Nump 1.14.5 .Python 3.6 .Cudq7.5+*For GPU .nltk 3.4 .tqdm 4.32.1 . Kenlm (https://github.com/kpu/kenlm)
Challenges we ran into
When we started to build we have problems to find out reasrch papers deal with great libraries suh as numpy cuda but we do that for pytorch users.
Accomplishments that we're proud of
We are proud that we worked with such great libraries numpy1.14.5, pytorch cuda and kenlm its was great experience and we learned alot..
What we learned
We learned different way and how to use such great libto make good use ful models liks TEXTGAN PYTORCH for pytorch, python influncers.
What's next for TEXTGAN PYTORCH
Its tha part of university project...
Log in or sign up for Devpost to join the conversation.