You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
After syncing the buckets, the final step is to define the different categories of entities that the named entity recognition model will be trained to predict. A configuration file is used to define the classes and to initialize the UI components to aid a user label entities. A sample config file has the following tags:
@@ -123,11 +124,11 @@ For general information, [visit LabelStudios templates page.](https://labelstud.
123
124
### **Labeling**
124
125
125
126
1. Select the task with **global_index** of 0, the **global index** indicates this is the start of the article and start labelling each task by moving onto the next **global_index** number.
126
-

127
+

127
128
-**Ensure pre-labelled entities are correct and/or fix:** we have tried to auto-tag entities to make this faster but it’s not perfect and this is what we’re improving, so this commonly misses entities or gets them partially right.
128
129
-**Label any missed entities:** these can be things with typos, words being smushed together, etc.
0 commit comments