renamed rules, script for counting rules, counted them in full corpus

kornai · Apr 11, 2017 · c6361f4 · c6361f4
1 parent 0675155
commit c6361f4
Show file tree

Hide file tree

Showing 4 changed files with 199 additions and 95 deletions.
diff --git a/exp/alto/output/en_ud_train_rule_count.txt b/exp/alto/output/en_ud_train_rule_count.txt
@@ -0,0 +1,105 @@
+  43010 _nbar1
+  40783 _nbar2
+  20427 _pos_n_noun
+  12120 _pos_v_verb
+   9915 _pos_n_propn
+   8901 _case_n_adp
+   8525 _det
+   8121 _pos_n_pron
+   7615 _vbar1
+   7327 _nsubj_v_n
+   6671 _punct_v2
+   6278 _pos_d_adp
+   5957 _obj
+   5734 _vbar2
+   5286 _punct_n2
+   5268 _root_v
+   4902 _obl_v_n
+   4464 _pos_a_adj
+   4440 _compound
+   4434 _amod
+   3883 _advmod_v_adv
+   3842 _aux
+   3570 _nmod
+   3312 _mark_v_sconj
+   3265 _root_n
+   2166 _nmod_poss
+   2121 _nsubj_n_n
+   1957 _pos_n_adj
+   1892 _advcl_v_v
+   1771 _cc_v
+   1715 _conj_n_n
+   1642 _cc_n
+   1555 _conj_v_v
+   1530 _pos_v_adj
+   1389 _pos_n_num
+   1385 _xcomp_v_v
+   1372 _ccomp
+   1333 _x_v2
+   1294 _x_n2
+   1204 _pos_d_pron
+   1201 _x1_n
+   1188 _cop_n
+   1175 _cop_a
+   1031 _punct_v1
+   1030 _nummod_n
+    987 _advmod_n_adv
+    971 _acl_relcl_n_v
+    835 _flat
+    777 _aux_pass
+    774 _appos
+    756 _acl
+    722 _parataxis
+    651 _nsubj_pass_v_n
+    531 _advmod_adj_adv
+    478 _discourse
+    440 _case_n_v
+    438 _advmod_v_part
+    431 _compound_prt
+    430 _xcomp_v_n
+    347 _obl_tmod
+    320 _conj_n_v
+    297 _nummod_num
+    275 _mark_n_sconj
+    239 _punct_n1
+    238 _expl_v
+    237 _x_intj
+    236 _advmod_adv_adv
+    232 _iobj
+    226 _nmod_tmod
+    206 _mark_adv
+    188 _list
+    160 _x1_v
+    153 _x1_adp
+    137 _obl_npmod
+    137 _conj_v_n
+    126 _mark_n_part
+    117 _pos_d_a
+    104 _nsubj_n_v
+     96 _root_x
+     96 _root_d
+     95 _advcl_v_n
+     91 _fixed
+     87 _obl_adv_n
+     80 _det_predet
+     74 _nmod_npmod
+     65 _pos_d_punct
+     60 _x_adv
+     55 _csubj_v_v
+     53 _x_sym
+     44 _advmod_n_part
+     35 _goeswith
+     30 _expl_a
+     29 _x_adp
+     17 _x_sconj
+     17 _mark_v_part
+     17 _cop_d
+     11 _csubj_n_v
+      6 _orphan
+      6 _acl_relcl_d_v
+      5 _x_cconj
+      5 _csubj_v_n
+      3 _nsubj_pass
+      1 _xcomp_v_adj
+      1 _pos_d_num
+      1 _csubj_pass_v_v
diff --git a/exp/alto/tools/count_rules.sh b/exp/alto/tools/count_rules.sh
@@ -0,0 +1,2 @@
+#!/usr/bin/env bash
+grep -v '^#' | grep -v '^$' | grep '^_null_$' -v | tr '(' '\n' | tr -s ')' '\n' | tr -d ','  | grep '^_' | sort | uniq -c | sort -nr