3 ms·
This looks great. I'd be curious to see the decision trees, and the training corpus/objective for them.
by AlexCoventry 8y ago
This looks great. I'd be curious to see the decision trees, and the training corpus/objective for them.
- addcn 8y agoThank you. You can find the code for that here: https://github.com/opticdev/marvin https://github.com/opticdev/marvin The objective of the training is bootstrappig a prettier. String template (pattern in the code) for each node type. So we can learn things like arguments are separated by “, “ but if the total # of chars of all of them is greater than X it’s more common to use “,\n”. Stuff like that. We had LSTMs for a while but they performed worse than a decision tree with a few key features as input.
- AlexCoventry 8y agoThanks, sounds very interesting. I'll take a look.