renamed gdtools repo to gd_tools

brown_gd_to_conll.py

@@ -7,8 +7,8 @@
 import re
 import sys
 from collections import namedtuple
-from gdtools.acainn import Lemmatizer
-from gdtools.acainn import Features
+from gd_tools.acainn import Lemmatizer
+from gd_tools.acainn import Features
 from pyconll.unit import Conll
 
 Split = namedtuple("split", "form1 upos1 xpos1 form2 upos2 xpos2")

brown_gd_to_dot_ccg.py

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 import pickle
 import sys
-from gdtools.acainn import Lemmatizer, Retagger, Subcat, Typer
+from gd_tools.acainn import Lemmatizer, Retagger, Subcat, Typer
 
 def tidy_word(string):
     """outputs string suitable for XMLification further down the pipeline"""

checker.py

@@ -1,6 +1,6 @@
 import numpy as np
 import pandas as pd
-from gdtools.acainn import Morphology
+from gd_tools.acainn import Morphology
 
 class Checker():
     # for simple matches

fix_feats.py

@@ -1,7 +1,7 @@
 import re
 import sys
 import pyconll
-from gdtools.acainn import Features
+from gd_tools.acainn import Features
 
 f = Features()
 

lemmatise.py

@@ -1,7 +1,7 @@
 """Overwrites the lemmata in a CoNLL-U file based on the form and XPOS."""
 import sys
 import pyconll
-from gdtools.acainn import Lemmatizer
+from gd_tools.acainn import Lemmatizer
 
 corpus = pyconll.load_from_file(sys.argv[1])
 l = Lemmatizer()

readme.md

@@ -19,6 +19,9 @@ In practice I have postprocessed the results with the following Python 3 scripts
 There is one small test tree bank in `ud`:
 * `gd_iomasgladh-ud-test.conllu` is a hand-built corpus from 2014 which has been converted to UD.
 
+The lemmatiser, code to convert ARCOSG parts of speech to UD features and categorial grammar code are now in the https://github.com/colinbatchelor/gd_tools repository.
+
+
 Earlier work
 --
 ### gramaran
@@ -37,8 +40,6 @@ Each sentence has three lines beginning with hashes preceding it. These are an I
 
 The guidelines used for the construction of the corpus in LaTeX format. Currently no special packages are used for it.
 
-
-
 * `brown_gd_to_dot_ccg.py` takes a Brown-format corpus assuming ARCOSG tags and outputs a .ccg file
 * `mend_xml.py` fixes the output of OpenCCG's ccg2xml.
 * `prepareARCOSG.py` takes a local installation of the Annotated Reference Corpus of Scottish Gaelic (ARCOSG), replaces spaces within tokens with underscores and puts the results in `arcosg.pkl`.
@@ -61,5 +62,5 @@ The citation for the material in `ccg` and `gramaran` is:
 
 Colin Batchelor
 
-2024-02-06
+2024-02-07
 

test_checker.py

@@ -1,5 +1,5 @@
 import unittest
-from gdtools.acainn import Morphology
+from gd_tools.acainn import Morphology
 from checker import Checker
 import numpy as np