search

mailgun / talon

Apache License 2.0
1.26k stars 287 forks source link

Training Talon with our email set #173

Closed qxh5696 closed 5 years ago

qxh5696 commented 5 years ago

Hey Mailgun Team,

How can I go about using our collection of emails to better train talon?

More specifically, what steps do I have to do to put our emails in format that can be fed into the training data.

I was reading this comment on an earlier issue and saw the following python file:

https://github.com/mailgun/talon/blob/master/talon/signature/learning/featurespace.py#L15

Do you mind posting an example of how to use this file to turn emails in plaintext into the format that would allow us to train talon on our own data sets?

obukhov-sergey commented 5 years ago

Fixed with https://github.com/mailgun/talon/pull/177

qxh5696 commented 5 years ago

Thanks for the update! Just to clarify,

"/path/to/talon/signature/data/classifier"

This is where the classifier that we are training in this example would be saved right?