add_nverb_entries.py
9.68 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
#-*- coding:utf-8 -*-
import codecs
from operator import itemgetter
from django.core.management.base import BaseCommand
from dictionary.models import Lemma, Lemma_Status, POS, Vocabulary, \
get_or_create_entry
from dictionary.management.commands.load_initial_nverb_frames import add_initial_frames_by_entries
from dictionary.management.commands.load_entries_relations import add_relations_by_nverb_entries
VERBS_IN_DICT = 2000
ORDERING = '300M'
################# NOUNS ################################
#VERBAL_NOUNS_PATH = 'data/nverbs/nouns/merged_nouns-freq.txt'
#NOUNS_VAL_PATH = 'data/nverbs/nouns/merged_nouns_val-poss.txt'
# loading initial entries
NOUNS_ADDED_PATH = 'data/nverbs/nouns2consider/added-merged_nouns_val_20171102.txt'
NOUNS_ERROR_PATH = 'data/nverbs/nouns2consider/error-merged_nouns_val_20171102.txt'
NOUNS_FRAMES_PATH = 'data/nverbs/nouns2consider/merged_nouns_val-poss.txt' # hasla z ramkami walencyjnymi z tajnego
SELECTED_NOUNS_PATH = 'data/nverbs/nouns2consider/nouns+verb2consider-clarin2-add.txt'
# adding entries relations
NOUN_VERB_RELATIONS_PATH = 'data/nverbs/nouns2consider/nouns+verb2consider-clarin2-add.txt'
NOUN_ADJ_RELATIONS_PATH = 'data/nverbs/nouns2consider/nouns+adj2consider-clarin2.txt'
################## ADJS ################################
#VERBAL_ADJS_PATH = 'data/nverbs/adjs/merged_adjs-freq.txt'
##ADJS_VAL_PATH = 'data/nverbs/adjs/merged_adjs_val-P1.txt'
#ADJS_RELATIONS_PATH = 'data/nverbs/adjs/ver_adjs+verb-freq_cuted.txt'
#
## loading initial entries
#ADJS_ADDED_PATH = 'data/nverbs/adjs/added-merged_adjs_val_20141219.txt'
#ADJS_ERROR_PATH = 'data/nverbs/adjs/error-merged_adjs_val_20141219.txt'
#ADJS_FRAMES_PATH = 'data/nverbs/adjs/merged_adjs_val-P1.txt'
#
## adding entries relations
#ADJ_VERB_RELATIONS_PATH = 'data/nverbs/adjs/merged_adjs+verb-freq.txt'
#
################## ADVS ################################
#VERBAL_ADVS_PATH = 'data/nverbs/advs/merged_advs-sel-1M-300M.txt' # frekwencja tylko wybrana
#ADVS_VAL_PATH = 'data/nverbs/advs/merged_advs_val_popr_usu_gdyby_20141113.txt'
#
## loading initial entries
#ADVS_ADDED_PATH = 'data/nverbs/advs/added-merged_advs_val.txt'
#ADVS_ERROR_PATH = 'data/nverbs/advs/error-merged_advs_val.txt'
#ADVS_FRAMES_PATH = 'data/nverbs/advs/merged_advs_val_popr_usu_gdyby_20141113.txt'
## adding entries relations # na razie brak danych
#ADV_VERB_RELATIONS_PATH = 'data/nverbs/adjs/merged_adjs+verb-freq.txt'
class Command(BaseCommand):
args = 'none'
def handle(self, **options):
# load nouns
# entries_with_val = get_entries(NOUNS_VAL_PATH)
# entries = get_entries_by_freq(VERBAL_NOUNS_PATH, ORDERING)
# load_entries(entries, B_entries, 'data/added_nouns_20140627.txt', ORDERING, 'noun',
# 'clarin_nouns', 1, 1, 0)
# load nouns
entries_to_add = get_entries(SELECTED_NOUNS_PATH)
ordered_entries = get_entries_by_freq(SELECTED_NOUNS_PATH, ORDERING)
#related_entries = get_related_entries(NOUNS_RELATIONS_PATH, 'noun')
added_entries = load_entries(ordered_entries, entries_to_add, 'data/nverbs/nouns2consider/added_nouns_20171103.txt', ORDERING, 'noun',
'clarin2_nouns', 3, 3, 0)
add_initial_frames_by_entries(added_entries,
NOUNS_FRAMES_PATH, NOUNS_ADDED_PATH, NOUNS_ERROR_PATH,
'noun')
add_relations_by_nverb_entries(added_entries, NOUN_VERB_RELATIONS_PATH, 'verb', 'noun')
#add_relations_by_nverb_entries(added_entries, NOUN_ADJ_RELATIONS_PATH, 'adj', 'noun')
# # load adjectives
## entries_with_val = get_entries(ADJS_VAL_PATH)
# entries = get_entries_by_freq(VERBAL_ADJS_PATH, ORDERING)
# related_entries = get_related_entries(ADJS_RELATIONS_PATH, 'adj')
# added_entries = load_entries(entries, related_entries, 'data/added_adjs_20141219.txt', ORDERING, 'adj',
# 'clarin_adjs', 3, 3, 0)
# add_initial_frames_by_entries(added_entries,
# ADJS_FRAMES_PATH, ADJS_ADDED_PATH, ADJS_ERROR_PATH,
# 'adj')
# add_relations_by_nverb_entries(added_entries, ADJ_VERB_RELATIONS_PATH, 'adj')
# # load adverbs
# entries_with_val = get_entries(ADVS_VAL_PATH)
# entries = get_entries_by_freq(VERBAL_ADVS_PATH, ORDERING)
# added_entries = load_entries(entries, entries_with_val, 'data/added_advs_20141114.txt', ORDERING, 'adv',
# 'clarin_advs', 1, 1, 0)
# add_initial_frames_by_entries(added_entries,
# ADVS_FRAMES_PATH, ADVS_ADDED_PATH, ADVS_ERROR_PATH,
# 'adverb')
# add_relations_by_nverb_entries(added_entries, ADJ_VERB_RELATIONS_PATH, 'adj')
def get_entries(entries_path):
entries = []
try:
entries_file = codecs.open(entries_path, "rt", 'utf-8')
for line in entries_file:
line_ls = line.split('\t')
entries.append({'entry' : line_ls[0].strip()})
entries = list(set(entries))
finally:
entries_file.close()
return entries
def get_entries_by_freq(entries_path, ordering):
entries = []
try:
freq_file = codecs.open(entries_path, "rt", 'utf-8')
for line in freq_file:
line_ls = line.split()
entries.append({'entry' : line_ls[0].strip(),
'freq_1M': int(line_ls[1].strip()),
'freq_300M': int(line_ls[2].strip())})
entries = sorted(entries, key=itemgetter('freq_%s' % ordering), reverse=True)
finally:
freq_file.close()
return entries
def load_entries(sorted_entries, entries_to_add, added_path, ordering, pos_tag,
dict_basename, first_dict_idx, last_dict_idx, min_freq):
print 'Loading entries!!'
try:
added_entries = []
added_file = codecs.open(added_path, "wt", 'utf-8')
dict = first_dict_idx
new_last_dict = last_dict_idx
pos_obj = POS.objects.get(tag=pos_tag)
verbs_per_dict = VERBS_IN_DICT
initial_status = Lemma_Status.objects.order_by('priority')[0]
for entry in sorted_entries:
found_entry = next((item for item in entries_to_add if item['entry'] == entry['entry']), None)
if found_entry and entry['freq_%s' % ordering] >= min_freq:
if dict == 0:
new_voc = Vocabulary(name=dict_basename)
new_voc.save()
else:
new_voc = Vocabulary(name=dict_basename+str(dict))
new_voc.save()
lemmas = Lemma.objects.filter(entry = entry['entry'])
if lemmas.count() == 0:
if verbs_per_dict == 0:
verbs_per_dict = VERBS_IN_DICT
dict += 1
if dict > new_last_dict:
break
else:
new_voc = Vocabulary(name=dict_basename+str(dict))
new_voc.save()
val_entry, created = get_or_create_entry(entry['entry'], pos_obj)
new_lemma = Lemma(entry=entry['entry'],
entry_obj=val_entry,
vocabulary=new_voc,
status=initial_status,
old=False,
frequency_300M=entry['freq_300M'],
frequency_1M=entry['freq_1M'])
new_lemma.save()
verbs_per_dict -= 1
added_file.write('%s\t%s\t%s\n' % (entry['entry'],
entry['freq_1M'],
entry['freq_300M']))
added_entries.append(entry['entry'])
print entry
finally:
added_file.close()
return added_entries
def get_related_entries(relations_path, pos_tag):
print 'Checking relations!!!'
entries = []
pos = POS.objects.get(tag=pos_tag)
try:
relations_file = codecs.open(relations_path, "rt", 'utf-8')
for line in relations_file:
#print line
line_ls = line.split()
verb = line_ls[3].lstrip('(').strip()
nverb = line_ls[0].strip()
if (not Lemma.objects.filter(entry=nverb).exists() and
Lemma.objects.filter(entry=verb, entry_obj__pos__tag='verb').exists()):
#entries.append(nverb)
verb_active_lemma = Lemma.objects.get(entry=verb,
entry_obj__pos__tag='verb',
old=False)
lemma_status_str = verb_active_lemma.status.status
if (not lemma_status_str == u'w obróbce' and
not lemma_status_str == u'do obróbki'):
if (verb_active_lemma.frames.count() == 1 and
verb_active_lemma.frames.filter(text_rep=u'subj{np(str)}').exists()):
pass
else:
entries.append({'entry' : nverb,
'verb' : verb,
'freq_1M': int(line_ls[1].strip()),
'freq_300M': int(line_ls[2].strip())})
print line
finally:
relations_file.close()
return entries