GitHub - anwang51/spaghettinet: Character-level LSTM trained on recipe database

anwang51 / spaghettinet Public

Notifications You must be signed in to change notification settings
Fork 0
Star 0

Character-level LSTM trained on recipe database

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 2 Commits
epicurious-recipes-with-rating-and-nutrition		epicurious-recipes-with-rating-and-nutrition
ReadMe.txt		ReadMe.txt
epicurious-recipes-with-rating-and-nutrition.zip		epicurious-recipes-with-rating-and-nutrition.zip
firstEpochState.h5		firstEpochState.h5
input.txt		input.txt
lstm_text_generation.py		lstm_text_generation.py
out.txt		out.txt
output1.txt		output1.txt
secondEpochBadLoss.h5		secondEpochBadLoss.h5

Repository files navigation

Run lstm_text_generation.py with python3 in interactive mode and call the function generateText() to produce some randomly generated text with the current weights.

The generatedText is written to out.txt, some generated text is already written in there.

The weights are the result of training over only one epoch, feel free to train this yourself but there seems to be some problem with either loading the weights or training the current weights as the loss function seems to steadily increase in the second epoch. It would probably be better to retrain the net (4~ hours per epoch).