How do I properly format txt to be used in deep learning text generation.
1 vue (au cours des 30 derniers jours)
Afficher commentaires plus anciens
Im currently following the 'Generate Text using Deep Learning' toolbox but using a different piece of text.
I don't understand where this part of the code comes from:
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/210454/image.png)
I understand what it does as i can see it in the text, but where does \x2403 come from. The reason i ask is because in my text, everywhere there is an apostrophe, whethere in a word like can't, or where theres are quotes this symbol shows up Ô ...
Later on when i try and train, i get this error:
Error using trainNetwork (line 165)
Invalid training data. Labels must not contain undefined values.
Error in txtgen (line 73)
net = trainNetwork(XTrain,YTrain,layers,options);
Im not sure if this is related but either way the Ô shouldn't be there i dont think...
0 commentaires
Réponses (1)
Harshit Jain
le 29 Mar 2019
Values of the form (\x0002) are unicode values for the respective characters. You can read more about unicode characters here
0 commentaires
Voir également
Catégories
En savoir plus sur Modeling and Prediction dans Help Center et File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!