Difference between revisions of "P2PaLATrainParameters"

From Transkribus Wiki
Jump to: navigation, search
Line 1: Line 1:
 
===Training Parameters for the P2PaLA structure tool (under construction)===
 
===Training Parameters for the P2PaLA structure tool (under construction)===
 
====Structure types====
 
====Structure types====
These are the structure types that are tagged using Transkriubs on <em>region</em> level. Do ''not'' use whitespace in those structure types and be careful with case sensitivity, i.e. we recommend using only lowercase letters. Also, don't use special characters like ! or ", although it may work...
+
These are the structure types that are tagged using Transkriubs on <em>region</em> level. Do ''not'' use whitespace in those structure types and be careful with case sensitivity, i.e. we recommend using only lowercase letters. Also we recommend to use dashes (-) and underscores (_) as the only special character, although other may work too.
 +
 
 +
Example:
 +
    paragraph heading footnote page-number
  
 
====Merged structure types====
 
====Merged structure types====
If you want to treat certain structure types like others (e.g. 'footnote-continued' or 'footer' like 'footnote') you can specify it here. Expected is a list of the structure types where other are merge into, separated by a colon, e.g.: 'footnote:footnote-continued,footer heading:header' whould mean that 'footnote-continue' and 'footnote' are regarded as 'footnote' while 'header' is regarded as 'heading'
+
Merged structure types are used to treat certain structure types the same as others during training (e.g. 'footnote-continued' or 'footer' like 'footnote'). Expected is a list of the structure types, separated by a colon with the structure types to merge.
 +
 
 +
Example:
 +
    footnote:footnote-continued,footer heading:header
 +
Here, regions tagged with 'footnote-continue' and 'footer' are regarded as 'footnote' while 'header' is regarded as 'heading' during training.
 +
 
  
 
For more information on P2PaLA, see also: https://transkribus.eu/wiki/index.php?title=P2PaLA
 
For more information on P2PaLA, see also: https://transkribus.eu/wiki/index.php?title=P2PaLA

Revision as of 07:34, 30 October 2019

Training Parameters for the P2PaLA structure tool (under construction)

Structure types

These are the structure types that are tagged using Transkriubs on region level. Do not use whitespace in those structure types and be careful with case sensitivity, i.e. we recommend using only lowercase letters. Also we recommend to use dashes (-) and underscores (_) as the only special character, although other may work too.

Example:

   paragraph heading footnote page-number

Merged structure types

Merged structure types are used to treat certain structure types the same as others during training (e.g. 'footnote-continued' or 'footer' like 'footnote'). Expected is a list of the structure types, separated by a colon with the structure types to merge.

Example:

   footnote:footnote-continued,footer heading:header

Here, regions tagged with 'footnote-continue' and 'footer' are regarded as 'footnote' while 'header' is regarded as 'heading' during training.


For more information on P2PaLA, see also: https://transkribus.eu/wiki/index.php?title=P2PaLA