Skip to content

Text-generating Neural Network, trained with GPU in Google Colab using Pytorch and word embeddings

Notifications You must be signed in to change notification settings

AbdelrahimKoura/StoryTeller

 
 

Repository files navigation

StoryTeller

Created by: Mariya Shavshyshvili
email: mariyasha888@gmail.com

Text-generating Neural Network, trained on Fairytales by The Brothers Grimm and Alice's Adventures in Wonderland by Lewis Carroll.
GPU connection is highly recomended for taining, as it may take over 10 hours to train on CPU.

Text files were generated from:
Fairytales: https://www.gutenberg.org/files/2591/2591-h/2591-h.htm
Alice: http://www.gutenberg.org/files/11/11-0.txt

IMPORTANT NOTES
Please replace the following variables in the notebook with the .txt files in the repository,
as this notebook links to my own Google Drive.
fairytales_url : ensure it links to FairytalesByTheBrothersGrimm.txt
alice_url : ensure it links to AlicesAdvanturesInWonderland.txt
checkpoint_url : any location to save model's checkpoint after training

Got lots of help with the word embeddings from Pytorch:
https://pytorch.org/tutorials/beginner/nlp/word_embeddings_tutorial.html#exercise-computing-word-embeddings-continuous-bag-of-words
Extra help regarding Google Colab's GPU connection:
https://www.kdnuggets.com/2018/02/fast-ai-lesson-1-google-colab-free-gpu.html

About

Text-generating Neural Network, trained with GPU in Google Colab using Pytorch and word embeddings

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 100.0%