search

Living-with-machines / DeezyMatch

A Flexible Deep Learning Approach to Fuzzy String Matching
https://living-with-machines.github.io/DeezyMatch/
Other
139 stars 34 forks source link

Issue with the vocab and finetuning #25

Closed kasra-hosseini closed 4 years ago

kasra-hosseini commented 4 years ago

Branch: https://github.com/Living-with-machines/DeezyMatch/tree/feature/25-vocab-issue

kasra-hosseini commented 4 years ago

This should be solved in this PR (under review): https://github.com/Living-with-machines/DeezyMatch/pull/26

mcollardanuy commented 4 years ago

Hi @fedenanni and @kasra-hosseini, if I'm not wrong, at the moment we use the vocabulary of the dataset of the original model for fine-tuning, so we fixed this by making sure we had a general vocabulary that covers both the original and potential fine-tuning vocabularies, when training the original model. This general vocabulary is hard-coded into the code, should we reopen this issue or create a new one?

kasra-hosseini commented 4 years ago

Good point. We can reopen this issue.

kasra-hosseini commented 4 years ago

PR https://github.com/Living-with-machines/DeezyMatch/pull/36 closes this issue.