KoichiYasuoka commited on
Commit
d26baab
1 Parent(s): 2abab5e
Files changed (1) hide show
  1. ud.py +7 -7
ud.py CHANGED
@@ -1,15 +1,15 @@
1
  from transformers import TokenClassificationPipeline
2
 
3
  class UniversalDependenciesPipeline(TokenClassificationPipeline):
4
- def _forward(self,model_input):
5
  import torch
6
- v=model_input["input_ids"][0].tolist()
7
  with torch.no_grad():
8
  e=self.model(input_ids=torch.tensor([v[0:i]+[self.tokenizer.mask_token_id]+v[i+1:]+[j] for i,j in enumerate(v[1:-1],1)]))
9
- return {"logits":e.logits[:,1:-2,:],**model_input}
10
- def postprocess(self,model_output,**kwargs):
11
  import numpy
12
- e=model_output["logits"].numpy()
13
  r=[1 if i==0 else -1 if j.endswith("|root") else 0 for i,j in sorted(self.model.config.id2label.items())]
14
  e+=numpy.where(numpy.add.outer(numpy.identity(e.shape[0]),r)==0,0,numpy.nan)
15
  g=self.model.config.label2id["X|_|goeswith"]
@@ -25,7 +25,7 @@ class UniversalDependenciesPipeline(TokenClassificationPipeline):
25
  k,h=z[numpy.nanargmax(m[z,z])],numpy.nanmin(m)-numpy.nanmax(m)
26
  m[:,z]+=[[0 if j in z and (i!=j or i==k) else h for i in z] for j in range(m.shape[0])]
27
  h=self.chu_liu_edmonds(m)
28
- v=[(s,e) for s,e in model_output["offset_mapping"][0].tolist() if s<e]
29
  q=[self.model.config.id2label[p[j,i]].split("|") for i,j in enumerate(h)]
30
  g="aggregation_strategy" in kwargs and kwargs["aggregation_strategy"]!="none"
31
  if g:
@@ -34,7 +34,7 @@ class UniversalDependenciesPipeline(TokenClassificationPipeline):
34
  h=[b if i>b else b-1 for a,b in enumerate(h) if i!=a]
35
  v[i-1]=(v[i-1][0],v.pop(i)[1])
36
  q.pop(i)
37
- t=model_output["sentence"].replace("\n"," ")
38
  u="# text = "+t+"\n"
39
  for i,(s,e) in enumerate(v):
40
  u+="\t".join([str(i+1),t[s:e],t[s:e] if g else "_",q[i][0],"_","|".join(q[i][1:-1]),str(0 if h[i]==i else h[i]+1),q[i][-1],"_","_" if i+1<len(v) and e<v[i+1][0] else "SpaceAfter=No"])+"\n"
1
  from transformers import TokenClassificationPipeline
2
 
3
  class UniversalDependenciesPipeline(TokenClassificationPipeline):
4
+ def _forward(self,model_inputs):
5
  import torch
6
+ v=model_inputs["input_ids"][0].tolist()
7
  with torch.no_grad():
8
  e=self.model(input_ids=torch.tensor([v[0:i]+[self.tokenizer.mask_token_id]+v[i+1:]+[j] for i,j in enumerate(v[1:-1],1)]))
9
+ return {"logits":e.logits[:,1:-2,:],**model_inputs}
10
+ def postprocess(self,model_outputs,**kwargs):
11
  import numpy
12
+ e=model_outputs["logits"].numpy()
13
  r=[1 if i==0 else -1 if j.endswith("|root") else 0 for i,j in sorted(self.model.config.id2label.items())]
14
  e+=numpy.where(numpy.add.outer(numpy.identity(e.shape[0]),r)==0,0,numpy.nan)
15
  g=self.model.config.label2id["X|_|goeswith"]
25
  k,h=z[numpy.nanargmax(m[z,z])],numpy.nanmin(m)-numpy.nanmax(m)
26
  m[:,z]+=[[0 if j in z and (i!=j or i==k) else h for i in z] for j in range(m.shape[0])]
27
  h=self.chu_liu_edmonds(m)
28
+ v=[(s,e) for s,e in model_outputs["offset_mapping"][0].tolist() if s<e]
29
  q=[self.model.config.id2label[p[j,i]].split("|") for i,j in enumerate(h)]
30
  g="aggregation_strategy" in kwargs and kwargs["aggregation_strategy"]!="none"
31
  if g:
34
  h=[b if i>b else b-1 for a,b in enumerate(h) if i!=a]
35
  v[i-1]=(v[i-1][0],v.pop(i)[1])
36
  q.pop(i)
37
+ t=model_outputs["sentence"].replace("\n"," ")
38
  u="# text = "+t+"\n"
39
  for i,(s,e) in enumerate(v):
40
  u+="\t".join([str(i+1),t[s:e],t[s:e] if g else "_",q[i][0],"_","|".join(q[i][1:-1]),str(0 if h[i]==i else h[i]+1),q[i][-1],"_","_" if i+1<len(v) and e<v[i+1][0] else "SpaceAfter=No"])+"\n"