Hello Jupyter Community!
I’m very green when it comes to the use of Jupyter and Coding in general,
but I have been assigned to create a Machine Learning solution ragrading text classification.
I have found algorithms that seem to work, but have been unable to cross-validate because of the sheer size of training data.
The cross-validation ran for 48 hours, until I shut it down.
However, I tried testing it with a smaller amount of training data, and it seemed to give me statisfying results.
I was able to fit the algorithm with the large amount of training data however, and I want to utilize the full potential of all the
training data that was collected - just without cross-validating to test it first.
Now my question is:
I have an algorithm, I have fitted it with the appropriate training data:
“How can I now use it?”
In the most basic way possible, how can I utilize the algortihm on new data.
I have so much history of words that have a certain classification - these are fitted to the algorithm.
-How can I input new words and receive the most likely class - as well as a percentage certainty (if that’s possible)
Most tutorials stop after testing the accuracy (some using cross-validation), but never show algorithms in practice.
If you want me to share any code please say so, but I hope I described pretty well what I got so far - I just need help with the final (most important) step
I want to stress that I am very new to the world of coding and Jupyter, so I would appreciate the answers in a not too technical fashion
Thank you so much in advance!
Best regards
- Laurits