Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,49 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: cc-by-nc-sa-4.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-sa-4.0
|
| 3 |
+
language:
|
| 4 |
+
- cs
|
| 5 |
+
- de
|
| 6 |
+
pipeline_tag: text2text-generation
|
| 7 |
+
tags:
|
| 8 |
+
- code
|
| 9 |
+
- wikibase
|
| 10 |
+
- vital records
|
| 11 |
+
base_model: google/mt5-small
|
| 12 |
+
inference:
|
| 13 |
+
parameters:
|
| 14 |
+
max_new_tokens: 100
|
| 15 |
+
---
|
| 16 |
+
|
| 17 |
+
A model for connverting vital records to Wikibase. Trained on a transcription dataset obrained with HTR (<a href="https://gitlab.com/scripta/escriptorium">eScriptorium</a>) from 150 pages of the Jewish registry book <a href="https://vademecum.nacr.cz/vademecum/permalink?xid=8c9e996c-e65c-41a0-8cbc-f8fc8dc32468">HBMa 2501</a> (<a href="https://www.wikidata.org/wiki/Q108724909">Wikidata</a>). The dataset contains 1071 vital records, which were uploaded to <a href="https://matrika.wikibase.cloud">Wikibase Cloud</a> after processing using this model.
|
| 18 |
+
|
| 19 |
+
Based on <a href="https://huggingface.co/google/mt5-small">Google's mT5</a>.
|
| 20 |
+
|
| 21 |
+
<h3>Processing ditto marks</h3>
|
| 22 |
+
|
| 23 |
+
<pre>
|
| 24 |
+
mt5prefix = "dto"
|
| 25 |
+
upperCell = "18 Juny 1858 H. Barb. Kohn"
|
| 26 |
+
lowerCell = "dto dto H. Ros. Plowitz"
|
| 27 |
+
dataset_input = f"{mt5prefix}:{upperCell}|{lowerCell}"
|
| 28 |
+
dataset_target = "18 Juny 1858 H. Ros. Plowitz"
|
| 29 |
+
</pre>
|
| 30 |
+
|
| 31 |
+
<h3>Processing a table cell</h3>
|
| 32 |
+
|
| 33 |
+
<pre>
|
| 34 |
+
mt5prefix = "wbPre"
|
| 35 |
+
header = "Namen des Vaters mit Anführung des Standes, ob Familiant od. Gewerbsmann, u. seines Wohnortes, Schutzdom. Haus-Nr. und bei ehelichen Geburten seines Gubernial-Ehekonsenses Datum und Zahl"
|
| 36 |
+
cell = "Markus Gibianer, Lederhändler, S. d. Seligmann Gibianer u. der Magdalena geb. Motz"
|
| 37 |
+
dataset_input = f"{mt5prefix}:{header}|{cell}"
|
| 38 |
+
dataset_target = "[[<a href="https://matrika.wikibase.cloud/wiki/Property:P23">P23</a>:Markus Gibianer|<a href="https://matrika.wikibase.cloud/wiki/Property:P26">P26</a>:Lederhändler]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P27">P27</a>:Seligmann Gibianer]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P28">P28</a>:Magdalena Gibianer geb. Motz]]"
|
| 39 |
+
</pre>
|
| 40 |
+
|
| 41 |
+
<h3>Processing a whole record</h3>
|
| 42 |
+
|
| 43 |
+
<pre>
|
| 44 |
+
mt5prefix = "wbDef"
|
| 45 |
+
input_string = "[[<a href="https://matrika.wikibase.cloud/wiki/Property:P5">P5</a>:60]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P6">P6</a>:1865-05-05]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P7">P7</a>:H. Charlotte Löw]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P12">P12</a>|DD:23]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P12">P12</a>|MM:04]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P12">P12</a>|YY:1865]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P14">P14</a>|<a href="https://matrika.wikibase.cloud/wiki/Property:P8">P8</a>:1061/2]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P14">P14</a>:<a href="https://matrika.wikibase.cloud/wiki/Item:Q8">Q8</a>]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P15">P15</a>|DD:06]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P15">P15</a>|MM:05]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P15">P15</a>|YY:1865]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P16">P16</a>:<a href="https://matrika.wikibase.cloud/wiki/Item:Q22">Q22</a>]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P17">P17</a>:Friederika]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P20">P20</a>:4]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P21">P21</a>:<a href="https://matrika.wikibase.cloud/wiki/Item:Q13">Q13</a>]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P22">P22</a>:<a href="https://matrika.wikibase.cloud/wiki/Item:Q15">Q15</a>]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P23">P23</a>:Jakob Raubitschek|<a href="https://matrika.wikibase.cloud/wiki/Property:P26">P26</a>:Eisenhändler]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P27">P27</a>:Josef Raubitschek]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P28">P28</a>:Theresia Raubitschek geb. Latus]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P24">P24</a>:Pauline XXX]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P29">P29</a>:Markus Raudnitz]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P30">P30</a>:Rachel Raudnitz geb. Kuh]]"
|
| 46 |
+
dataset_input = f"{mt5prefix}:{input_string}"
|
| 47 |
+
dataset_target = "[[<a href="https://matrika.wikibase.cloud/wiki/Property:P10">P10</a>:<a href="https://matrika.wikibase.cloud/wiki/Item:Q5">Q5</a>]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P12">P12</a>:1865-04-23]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P14">P14</a>:<a href="https://matrika.wikibase.cloud/wiki/Item:Q8">Q8</a>|<a href="https://matrika.wikibase.cloud/wiki/Property:P8">P8</a>:1061/2]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P15">P15</a>:1865-05-06]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P16">P16</a>:<a href="https://matrika.wikibase.cloud/wiki/Item:Q22">Q22</a>]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P17">P17</a>:Friederika]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P18">P18</a>:Raubitschek]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P19">P19</a>:Friederika Raubitschek]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P1">P1</a>:<a href="https://matrika.wikibase.cloud/wiki/Item:Q1">Q1</a>]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P20">P20</a>:4]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P21">P21</a>:<a href="https://matrika.wikibase.cloud/wiki/Item:Q13">Q13</a>]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P22">P22</a>:<a href="https://matrika.wikibase.cloud/wiki/Item:Q15">Q15</a>]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P23">P23</a>:Jakob Raubitschek|<a href="https://matrika.wikibase.cloud/wiki/Property:P26">P26</a>:Eisenhändler]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P24">P24</a>:Pauline Raubitschek]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P27">P27</a>:Josef Raubitschek]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P28">P28</a>:Theresia Raubitschek geb. Latus]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P29">P29</a>:Markus Raudnitz]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P30">P30</a>:Rachel Raudnitz geb. Kuh]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P33">P33</a>:Charlotte Löw]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P38">P38</a>:<a href="https://matrika.wikibase.cloud/wiki/Item:Q170">Q170</a>]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P5">P5</a>:60]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P6">P6</a>:1865-05-05]] [[<a href="https://matrika.wikibase.cloud/wiki/Property:P7">P7</a>:H. Charlotte Löw]]"
|
| 48 |
+
</pre>
|
| 49 |
+
|