How to fine tune Giveme5W1H from learnt weights?

fhamborg / Giveme5W1H

Extraction of the journalistic five W and one H questions (5W1H) from news articles: who did what, when, where, why, and how?

Apache License 2.0

512 stars 89 forks source link

How to fine tune Giveme5W1H from learnt weights? #74

Open vblong opened 2 years ago

vblong commented 2 years ago

Describe the bug This is not actually a bug. I managed to run the sample code to produce weights. I wanted to know how can I use the result from learning weights from my own dataset to improve/fine-tune the system?

To Reproduce Follow the steps described in https://github.com/fhamborg/Giveme5W1H/blob/master/Giveme5W1H/examples/misc/learn_weights/README.md