Fixed bug with UNK tokens being discarded causing misalignment.
Browse files- BertForJointParsing.py +6 -4
- BertForMorphTagging.py +1 -0
- BertForPrefixMarking.py +48 -30
- BertForSyntaxParsing.py +1 -0
BertForJointParsing.py
CHANGED
@@ -199,7 +199,7 @@ class BertForJointParsing(BertPreTrainedModel):
|
|
199 |
|
200 |
# predict the logits for the sentence
|
201 |
if self.prefix is not None:
|
202 |
-
inputs = encode_sentences_for_bert_for_prefix_marking(tokenizer, sentences, padding)
|
203 |
else:
|
204 |
inputs = tokenizer(sentences, padding=padding, truncation=truncation, return_offsets_mapping=True, return_tensors='pt')
|
205 |
|
@@ -218,7 +218,7 @@ class BertForJointParsing(BertPreTrainedModel):
|
|
218 |
|
219 |
# Prefix logits: each sentence gets a list([prefix_segment, word_without_prefix]) - **WITH CLS & SEP**
|
220 |
if output.prefix_logits is not None:
|
221 |
-
for sent_idx,parsed in enumerate(prefix_parse_logits(input_ids, sentences, tokenizer, output.prefix_logits)):
|
222 |
merge_token_list(final_output[sent_idx]['tokens'], map(tuple, parsed[1:-1]), 'seg')
|
223 |
|
224 |
# Lex logits each sentence gets a list(tuple(word, lexeme))
|
@@ -272,6 +272,7 @@ def combine_token_wordpieces(input_ids: List[int], offset_mapping: torch.Tensor,
|
|
272 |
offset_mapping = offset_mapping.tolist()
|
273 |
ret = []
|
274 |
special_toks = tokenizer.all_special_tokens
|
|
|
275 |
for token, offsets in zip(tokenizer.convert_ids_to_tokens(input_ids), offset_mapping):
|
276 |
if token in special_toks: continue
|
277 |
if token.startswith('##'):
|
@@ -285,6 +286,7 @@ def ner_parse_logits(input_ids: List[List[int]], sentences: List[str], tokenizer
|
|
285 |
batch_ret = []
|
286 |
|
287 |
special_toks = tokenizer.all_special_tokens
|
|
|
288 |
for batch_idx in range(len(sentences)):
|
289 |
|
290 |
ret = []
|
@@ -311,6 +313,7 @@ def lex_parse_logits(input_ids: List[List[int]], sentences: List[str], tokenizer
|
|
311 |
batch_ret = []
|
312 |
|
313 |
special_toks = tokenizer.all_special_tokens
|
|
|
314 |
for batch_idx in range(len(sentences)):
|
315 |
intermediate_ret = []
|
316 |
tokens = tokenizer.convert_ids_to_tokens(input_ids[batch_idx])
|
@@ -519,5 +522,4 @@ def ud_get_prefix_dep(pre, word, word_idx):
|
|
519 |
if pre == 'ื':
|
520 |
func = 'det' if 'DET' in word['morph']['prefixes'] else 'mark'
|
521 |
|
522 |
-
return (word['syntax']['dep_head_idx'] if does_follow_main else word_idx), func
|
523 |
-
|
|
|
199 |
|
200 |
# predict the logits for the sentence
|
201 |
if self.prefix is not None:
|
202 |
+
inputs = encode_sentences_for_bert_for_prefix_marking(tokenizer, self.config.prefix_cfg, sentences, padding)
|
203 |
else:
|
204 |
inputs = tokenizer(sentences, padding=padding, truncation=truncation, return_offsets_mapping=True, return_tensors='pt')
|
205 |
|
|
|
218 |
|
219 |
# Prefix logits: each sentence gets a list([prefix_segment, word_without_prefix]) - **WITH CLS & SEP**
|
220 |
if output.prefix_logits is not None:
|
221 |
+
for sent_idx,parsed in enumerate(prefix_parse_logits(input_ids, sentences, tokenizer, output.prefix_logits, self.config.prefix_cfg)):
|
222 |
merge_token_list(final_output[sent_idx]['tokens'], map(tuple, parsed[1:-1]), 'seg')
|
223 |
|
224 |
# Lex logits each sentence gets a list(tuple(word, lexeme))
|
|
|
272 |
offset_mapping = offset_mapping.tolist()
|
273 |
ret = []
|
274 |
special_toks = tokenizer.all_special_tokens
|
275 |
+
special_toks.remove(tokenizer.unk_token)
|
276 |
for token, offsets in zip(tokenizer.convert_ids_to_tokens(input_ids), offset_mapping):
|
277 |
if token in special_toks: continue
|
278 |
if token.startswith('##'):
|
|
|
286 |
batch_ret = []
|
287 |
|
288 |
special_toks = tokenizer.all_special_tokens
|
289 |
+
special_toks.remove(tokenizer.unk_token)
|
290 |
for batch_idx in range(len(sentences)):
|
291 |
|
292 |
ret = []
|
|
|
313 |
batch_ret = []
|
314 |
|
315 |
special_toks = tokenizer.all_special_tokens
|
316 |
+
special_toks.remove(tokenizer.unk_token)
|
317 |
for batch_idx in range(len(sentences)):
|
318 |
intermediate_ret = []
|
319 |
tokens = tokenizer.convert_ids_to_tokens(input_ids[batch_idx])
|
|
|
522 |
if pre == 'ื':
|
523 |
func = 'det' if 'DET' in word['morph']['prefixes'] else 'mark'
|
524 |
|
525 |
+
return (word['syntax']['dep_head_idx'] if does_follow_main else word_idx), func
|
|
BertForMorphTagging.py
CHANGED
@@ -176,6 +176,7 @@ def parse_logits(input_ids: List[List[int]], sentences: List[str], tokenizer: Be
|
|
176 |
# Where tokens is a list of dicts, where each dict is:
|
177 |
# { pos: str, feats: dict, prefixes: List[str], suffix: str | bool, suffix_feats: dict | None}
|
178 |
special_toks = tokenizer.all_special_tokens
|
|
|
179 |
ret = []
|
180 |
for sent_idx,sentence in enumerate(sentences):
|
181 |
input_id_strs = tokenizer.convert_ids_to_tokens(input_ids[sent_idx])
|
|
|
176 |
# Where tokens is a list of dicts, where each dict is:
|
177 |
# { pos: str, feats: dict, prefixes: List[str], suffix: str | bool, suffix_feats: dict | None}
|
178 |
special_toks = tokenizer.all_special_tokens
|
179 |
+
special_toks.remove(tokenizer.unk_token)
|
180 |
ret = []
|
181 |
for sent_idx,sentence in enumerate(sentences):
|
182 |
input_id_strs = tokenizer.convert_ids_to_tokens(input_ids[sent_idx])
|
BertForPrefixMarking.py
CHANGED
@@ -7,18 +7,31 @@ from transformers import BertPreTrainedModel, BertModel, BertTokenizerFast
|
|
7 |
|
8 |
# define the classes, and the possible prefixes for each class
|
9 |
POSSIBLE_PREFIX_CLASSES = [ ['ืืืฉ', 'ืืฉ', 'ืืฉ', 'ืืฉ', 'ืืฉ'], ['ื'], ['ืฉ'], ['ื'], ['ื'], ['ื'], ['ื'], ['ื'] ]
|
10 |
-
|
11 |
-
|
12 |
-
|
13 |
-
#
|
14 |
-
|
15 |
-
|
16 |
-
|
17 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
# keep trimming prefixes from the string
|
19 |
-
while len(s) > 0 and s[0] in
|
20 |
# find the longest string to trim
|
21 |
-
next_pre = next((pre for pre in
|
22 |
if next_pre is None:
|
23 |
return
|
24 |
yield next_pre
|
@@ -30,9 +43,9 @@ def get_prefixes_from_str(s, greedy=False):
|
|
30 |
yield next_pre[0]
|
31 |
s = s[len(next_pre):]
|
32 |
|
33 |
-
def get_prefix_classes_from_str(s, greedy=False):
|
34 |
-
for pre in get_prefixes_from_str(s, greedy):
|
35 |
-
yield
|
36 |
|
37 |
@dataclass
|
38 |
class PrefixesClassifiersOutput(ModelOutput):
|
@@ -46,16 +59,21 @@ class BertPrefixMarkingHead(nn.Module):
|
|
46 |
super().__init__()
|
47 |
self.config = config
|
48 |
|
|
|
|
|
|
|
|
|
|
|
49 |
# an embedding table containing an embedding for each prefix class + 1 for NONE
|
50 |
# we will concatenate either the embedding/NONE for each class - and we want the concatenate
|
51 |
# size to be the hidden_size
|
52 |
-
prefix_class_embed = config.hidden_size //
|
53 |
-
self.prefix_class_embeddings = nn.Embedding(
|
54 |
|
55 |
# one layer for transformation, apply an activation, then another N classifiers for each prefix class
|
56 |
-
self.transform = nn.Linear(config.hidden_size + prefix_class_embed *
|
57 |
self.activation = nn.Tanh()
|
58 |
-
self.classifiers = nn.ModuleList([nn.Linear(config.hidden_size, 2) for _ in range(
|
59 |
|
60 |
def forward(
|
61 |
self,
|
@@ -66,8 +84,8 @@ class BertPrefixMarkingHead(nn.Module):
|
|
66 |
# encode the prefix_class_id_options
|
67 |
# If input_ids is batch x seq_len
|
68 |
# Then sequence_output is batch x seq_len x hidden_dim
|
69 |
-
# So prefix_class_id_options is batch x seq_len x
|
70 |
-
# Looking up the embeddings should give us batch x seq_len x
|
71 |
possible_class_embed = self.prefix_class_embeddings(prefix_class_id_options)
|
72 |
# then flatten the final dimension - now we have batch x seq_len x hidden_dim_2
|
73 |
possible_class_embed = possible_class_embed.reshape(possible_class_embed.shape[:-2] + (-1,))
|
@@ -148,15 +166,15 @@ class BertForPrefixMarking(BertPreTrainedModel):
|
|
148 |
|
149 |
def predict(self, sentences: List[str], tokenizer: BertTokenizerFast, padding='longest'):
|
150 |
# step 1: encode the sentences through using the tokenizer, and get the input tensors + prefix id tensors
|
151 |
-
inputs = encode_sentences_for_bert_for_prefix_marking(tokenizer, sentences, padding)
|
152 |
inputs.pop('offset_mapping')
|
153 |
inputs = {k:v.to(self.device) for k,v in inputs.items()}
|
154 |
|
155 |
# run through bert
|
156 |
logits = self.forward(**inputs, return_dict=True).logits
|
157 |
-
return parse_logits(inputs['input_ids'].tolist(), sentences, tokenizer, logits)
|
158 |
|
159 |
-
def parse_logits(input_ids: List[List[int]], sentences: List[str], tokenizer: BertTokenizerFast, logits: torch.FloatTensor):
|
160 |
# extract the predictions by argmaxing the final dimension (batch x sequence x prefixes x prediction)
|
161 |
logit_preds = torch.argmax(logits, axis=3).tolist()
|
162 |
|
@@ -176,7 +194,7 @@ def parse_logits(input_ids: List[List[int]], sentences: List[str], tokenizer: Be
|
|
176 |
token += tokens[next_tok_idx][2:]
|
177 |
next_tok_idx += 1
|
178 |
|
179 |
-
prefix_len = get_predicted_prefix_len_from_logits(token, logit_preds[sent_idx][tok_idx])
|
180 |
|
181 |
if not prefix_len:
|
182 |
ret[-1].append([token])
|
@@ -184,18 +202,18 @@ def parse_logits(input_ids: List[List[int]], sentences: List[str], tokenizer: Be
|
|
184 |
ret[-1].append([token[:prefix_len], token[prefix_len:]])
|
185 |
return ret
|
186 |
|
187 |
-
def encode_sentences_for_bert_for_prefix_marking(tokenizer: BertTokenizerFast, sentences: List[str], padding='longest', truncation=True):
|
188 |
inputs = tokenizer(sentences, padding=padding, truncation=truncation, return_offsets_mapping=True, return_tensors='pt')
|
189 |
# create our prefix_id_options array which will be like the input ids shape but with an addtional
|
190 |
# dimension containing for each prefix whether it can be for that word
|
191 |
-
prefix_id_options = torch.full(inputs['input_ids'].shape + (
|
192 |
|
193 |
# go through each token, and fill in the vector accordingly
|
194 |
for sent_idx, sent_ids in enumerate(inputs['input_ids']):
|
195 |
tokens = tokenizer.convert_ids_to_tokens(sent_ids)
|
196 |
for tok_idx, token in enumerate(tokens):
|
197 |
# if the first letter isn't a valid prefix letter, nothing to talk about
|
198 |
-
if len(token) < 2 or not token[0] in
|
199 |
|
200 |
# combine the next tokens in? only if it's a breakup
|
201 |
next_tok_idx = tok_idx + 1
|
@@ -204,13 +222,13 @@ def encode_sentences_for_bert_for_prefix_marking(tokenizer: BertTokenizerFast, s
|
|
204 |
next_tok_idx += 1
|
205 |
|
206 |
# find all the possible prefixes - and mark them as 0 (and in the possible mark it as it's value for embed lookup)
|
207 |
-
for pre_class in get_prefix_classes_from_str(token):
|
208 |
prefix_id_options[sent_idx, tok_idx, pre_class] = pre_class
|
209 |
|
210 |
inputs['prefix_class_id_options'] = prefix_id_options
|
211 |
return inputs
|
212 |
|
213 |
-
def get_predicted_prefix_len_from_logits(token, token_logits):
|
214 |
# Go through each possible prefix, and check if the prefix is yes - and if
|
215 |
# so increase the counter of the matched length, otherwise break out. That will solve cases
|
216 |
# of predicting prefix combinations that don't exist on the word.
|
@@ -221,7 +239,7 @@ def get_predicted_prefix_len_from_logits(token, token_logits):
|
|
221 |
# 2] Always check that the word starts with that prefix - otherwise it's bad
|
222 |
# (except for the case of multi-letter prefix, where we force the next to be last)
|
223 |
cur_len, skip_next, last_check, seen_prefixes = 0, False, False, set()
|
224 |
-
for prefix in get_prefixes_from_str(token):
|
225 |
# Are we skipping this prefix? This will be the case where we matched ืืฉ, don't allow ืฉ
|
226 |
if skip_next:
|
227 |
skip_next = False
|
@@ -232,7 +250,7 @@ def get_predicted_prefix_len_from_logits(token, token_logits):
|
|
232 |
seen_prefixes.add(prefix)
|
233 |
|
234 |
# check if we predicted this prefix
|
235 |
-
if token_logits[
|
236 |
cur_len += len(prefix)
|
237 |
if last_check: break
|
238 |
skip_next = len(prefix) > 1
|
|
|
7 |
|
8 |
# define the classes, and the possible prefixes for each class
|
9 |
POSSIBLE_PREFIX_CLASSES = [ ['ืืืฉ', 'ืืฉ', 'ืืฉ', 'ืืฉ', 'ืืฉ'], ['ื'], ['ืฉ'], ['ื'], ['ื'], ['ื'], ['ื'], ['ื'] ]
|
10 |
+
POSSIBLE_RABBINIC_PREFIX_CLASSES = [ ['ืืืฉ', 'ืืฉ', 'ืืฉ', 'ืืฉ', 'ืืฉ', 'ืื', 'ืื', 'ืื', 'ืื', 'ืืื'], ['ื'], ['ืฉ', 'ื'], ['ื'], ['ื'], ['ื'], ['ื'], ['ื'], ['ื'], ['ืง'] ]
|
11 |
+
|
12 |
+
class PrefixConfig(dict):
|
13 |
+
def __init__(self, possible_classes, **kwargs): # added kwargs for previous version where all features were kept as dict values
|
14 |
+
super().__init__()
|
15 |
+
self.possible_classes = possible_classes
|
16 |
+
self.total_classes = len(possible_classes)
|
17 |
+
self.prefix_c2i = {w: i for i, l in enumerate(possible_classes) for w in l}
|
18 |
+
self.all_prefix_items = list(sorted(self.prefix_c2i.keys(), key=len, reverse=True))
|
19 |
+
|
20 |
+
@property
|
21 |
+
def possible_classes(self) -> List[List[str]]:
|
22 |
+
return self.get('possible_classes')
|
23 |
+
|
24 |
+
@possible_classes.setter
|
25 |
+
def possible_classes(self, value: List[List[str]]):
|
26 |
+
self['possible_classes'] = value
|
27 |
+
|
28 |
+
DEFAULT_PREFIX_CONFIG = PrefixConfig(POSSIBLE_PREFIX_CLASSES)
|
29 |
+
|
30 |
+
def get_prefixes_from_str(s, cfg: PrefixConfig, greedy=False):
|
31 |
# keep trimming prefixes from the string
|
32 |
+
while len(s) > 0 and s[0] in cfg.prefix_c2i:
|
33 |
# find the longest string to trim
|
34 |
+
next_pre = next((pre for pre in cfg.all_prefix_items if s.startswith(pre)), None)
|
35 |
if next_pre is None:
|
36 |
return
|
37 |
yield next_pre
|
|
|
43 |
yield next_pre[0]
|
44 |
s = s[len(next_pre):]
|
45 |
|
46 |
+
def get_prefix_classes_from_str(s, cfg: PrefixConfig, greedy=False):
|
47 |
+
for pre in get_prefixes_from_str(s, cfg, greedy):
|
48 |
+
yield cfg.prefix_c2i[pre]
|
49 |
|
50 |
@dataclass
|
51 |
class PrefixesClassifiersOutput(ModelOutput):
|
|
|
59 |
super().__init__()
|
60 |
self.config = config
|
61 |
|
62 |
+
if not hasattr(config, 'prefix_cfg') or config.prefix_cfg is None:
|
63 |
+
setattr(config, 'prefix_cfg', DEFAULT_PREFIX_CONFIG)
|
64 |
+
if isinstance(config.prefix_cfg, dict):
|
65 |
+
config.prefix_cfg = PrefixConfig(config.prefix_cfg['possible_classes'])
|
66 |
+
|
67 |
# an embedding table containing an embedding for each prefix class + 1 for NONE
|
68 |
# we will concatenate either the embedding/NONE for each class - and we want the concatenate
|
69 |
# size to be the hidden_size
|
70 |
+
prefix_class_embed = config.hidden_size // config.prefix_cfg.total_classes
|
71 |
+
self.prefix_class_embeddings = nn.Embedding(config.prefix_cfg.total_classes + 1, prefix_class_embed)
|
72 |
|
73 |
# one layer for transformation, apply an activation, then another N classifiers for each prefix class
|
74 |
+
self.transform = nn.Linear(config.hidden_size + prefix_class_embed * config.prefix_cfg.total_classes, config.hidden_size)
|
75 |
self.activation = nn.Tanh()
|
76 |
+
self.classifiers = nn.ModuleList([nn.Linear(config.hidden_size, 2) for _ in range(config.prefix_cfg.total_classes)])
|
77 |
|
78 |
def forward(
|
79 |
self,
|
|
|
84 |
# encode the prefix_class_id_options
|
85 |
# If input_ids is batch x seq_len
|
86 |
# Then sequence_output is batch x seq_len x hidden_dim
|
87 |
+
# So prefix_class_id_options is batch x seq_len x total_classes
|
88 |
+
# Looking up the embeddings should give us batch x seq_len x total_classes x hidden_dim / N
|
89 |
possible_class_embed = self.prefix_class_embeddings(prefix_class_id_options)
|
90 |
# then flatten the final dimension - now we have batch x seq_len x hidden_dim_2
|
91 |
possible_class_embed = possible_class_embed.reshape(possible_class_embed.shape[:-2] + (-1,))
|
|
|
166 |
|
167 |
def predict(self, sentences: List[str], tokenizer: BertTokenizerFast, padding='longest'):
|
168 |
# step 1: encode the sentences through using the tokenizer, and get the input tensors + prefix id tensors
|
169 |
+
inputs = encode_sentences_for_bert_for_prefix_marking(tokenizer, self.config.prefix_cfg, sentences, padding)
|
170 |
inputs.pop('offset_mapping')
|
171 |
inputs = {k:v.to(self.device) for k,v in inputs.items()}
|
172 |
|
173 |
# run through bert
|
174 |
logits = self.forward(**inputs, return_dict=True).logits
|
175 |
+
return parse_logits(inputs['input_ids'].tolist(), sentences, tokenizer, logits, self.config.prefix_cfg)
|
176 |
|
177 |
+
def parse_logits(input_ids: List[List[int]], sentences: List[str], tokenizer: BertTokenizerFast, logits: torch.FloatTensor, config: PrefixConfig):
|
178 |
# extract the predictions by argmaxing the final dimension (batch x sequence x prefixes x prediction)
|
179 |
logit_preds = torch.argmax(logits, axis=3).tolist()
|
180 |
|
|
|
194 |
token += tokens[next_tok_idx][2:]
|
195 |
next_tok_idx += 1
|
196 |
|
197 |
+
prefix_len = get_predicted_prefix_len_from_logits(token, logit_preds[sent_idx][tok_idx], config)
|
198 |
|
199 |
if not prefix_len:
|
200 |
ret[-1].append([token])
|
|
|
202 |
ret[-1].append([token[:prefix_len], token[prefix_len:]])
|
203 |
return ret
|
204 |
|
205 |
+
def encode_sentences_for_bert_for_prefix_marking(tokenizer: BertTokenizerFast, config: PrefixConfig, sentences: List[str], padding='longest', truncation=True):
|
206 |
inputs = tokenizer(sentences, padding=padding, truncation=truncation, return_offsets_mapping=True, return_tensors='pt')
|
207 |
# create our prefix_id_options array which will be like the input ids shape but with an addtional
|
208 |
# dimension containing for each prefix whether it can be for that word
|
209 |
+
prefix_id_options = torch.full(inputs['input_ids'].shape + (config.total_classes,), config.total_classes, dtype=torch.long)
|
210 |
|
211 |
# go through each token, and fill in the vector accordingly
|
212 |
for sent_idx, sent_ids in enumerate(inputs['input_ids']):
|
213 |
tokens = tokenizer.convert_ids_to_tokens(sent_ids)
|
214 |
for tok_idx, token in enumerate(tokens):
|
215 |
# if the first letter isn't a valid prefix letter, nothing to talk about
|
216 |
+
if len(token) < 2 or not token[0] in config.prefix_c2i: continue
|
217 |
|
218 |
# combine the next tokens in? only if it's a breakup
|
219 |
next_tok_idx = tok_idx + 1
|
|
|
222 |
next_tok_idx += 1
|
223 |
|
224 |
# find all the possible prefixes - and mark them as 0 (and in the possible mark it as it's value for embed lookup)
|
225 |
+
for pre_class in get_prefix_classes_from_str(token, config):
|
226 |
prefix_id_options[sent_idx, tok_idx, pre_class] = pre_class
|
227 |
|
228 |
inputs['prefix_class_id_options'] = prefix_id_options
|
229 |
return inputs
|
230 |
|
231 |
+
def get_predicted_prefix_len_from_logits(token, token_logits, config: PrefixConfig):
|
232 |
# Go through each possible prefix, and check if the prefix is yes - and if
|
233 |
# so increase the counter of the matched length, otherwise break out. That will solve cases
|
234 |
# of predicting prefix combinations that don't exist on the word.
|
|
|
239 |
# 2] Always check that the word starts with that prefix - otherwise it's bad
|
240 |
# (except for the case of multi-letter prefix, where we force the next to be last)
|
241 |
cur_len, skip_next, last_check, seen_prefixes = 0, False, False, set()
|
242 |
+
for prefix in get_prefixes_from_str(token, config):
|
243 |
# Are we skipping this prefix? This will be the case where we matched ืืฉ, don't allow ืฉ
|
244 |
if skip_next:
|
245 |
skip_next = False
|
|
|
250 |
seen_prefixes.add(prefix)
|
251 |
|
252 |
# check if we predicted this prefix
|
253 |
+
if token_logits[config.prefix_c2i[prefix]]:
|
254 |
cur_len += len(prefix)
|
255 |
if last_check: break
|
256 |
skip_next = len(prefix) > 1
|
BertForSyntaxParsing.py
CHANGED
@@ -166,6 +166,7 @@ def parse_logits(input_ids: List[List[int]], sentences: List[str], tokenizer: Be
|
|
166 |
outputs = []
|
167 |
|
168 |
special_toks = tokenizer.all_special_tokens
|
|
|
169 |
for i in range(len(sentences)):
|
170 |
deps = logits.dependency_head_indices[i].tolist()
|
171 |
funcs = logits.function_logits.argmax(-1)[i].tolist()
|
|
|
166 |
outputs = []
|
167 |
|
168 |
special_toks = tokenizer.all_special_tokens
|
169 |
+
special_toks.remove(tokenizer.unk_token)
|
170 |
for i in range(len(sentences)):
|
171 |
deps = logits.dependency_head_indices[i].tolist()
|
172 |
funcs = logits.function_logits.argmax(-1)[i].tolist()
|