Answer the question
In order to leave comments, you need to log in
How to prepare the base for a neural network?
Good day! I want to understand how neural networks work and for this I want to write a program. It should just classify the text.
I decided to start with a dictionary for a neural network and questions arose
1) Where is it better to store the database?
2) How to store base?
You need to split texts using ngram or apply to the network input by splitting text using ngram
Thank you!
Answer the question
In order to leave comments, you need to log in
Have you read at least one book, article, or just a page on the Internet about neural networks, how they are arranged, how they work? Or did you decide to "reinvent the wheel"?
Judging by the questions - did not read. This is where I recommend starting.
About the database - the answer is about the same. Try to start by understanding how you can store information in a computer. It is written on the first pages of any book on programming.
About "ngram" - this is generally some kind of mysterious phrase. At least put some punctuation marks, or something.
Training data is usually stored simply as text files.
The result of training (any additional data, except for model files) - in any convenient form. You yourself will then use them in your services.
Didn't find what you were looking for?
Ask your questionAsk a Question
731 491 924 answers to any question