To reduce the character set, I encoded the text with unicode escapes Indeed, the file contains some 5300 unique characters-many of which appear only a handful of times-and I suspected that the character set was simply too large for the neural network to effectively learn on. My suspicion was that the training data had too many symbols. However, initial results were perhaps a little too vapor:Ī few more hours of training helped, but not as significantly as I hoped. My first attempt trained the neural network directly against the data scraped from Bandcamp. This is a character level recurrent neural network, which means that it learns pretty much just by looking at large samples of text. You can find the (horrific) script used to scrape the data here, along with the complete training file.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |