Closed saurabhg-predii closed 1 year ago
I am running the notebook on Colab on Python v3.9.16 Kindly help me resolve the issue @rodrigonogueira4, @jayralencar.
1 from pygaggle.rerank.base import Query, Text ----> 2 from pygaggle.rerank.transformer import MonoT5 3 4 reranker = MonoT5(model=model)
7 frames /usr/local/lib/python3.9/dist-packages/pygaggle/rerank/transformer.py in 11 import torch 12 from .base import Reranker, Query, Text ---> 13 from .similarity import SimilarityMatrixProvider 14 from pygaggle.model import (BatchTokenizer, 15 LongBatchEncoder,
/usr/local/lib/python3.9/dist-packages/pygaggle/rerank/similarity.py in 3 import torch 4 ----> 5 from pygaggle.model.encode import SingleEncoderOutput 6 7
/usr/local/lib/python3.9/dist-packages/pygaggle/model/init.py in 4 5 from .decode import ----> 6 from .encode import 7 from .evaluate import 8 from .serialize import
/usr/local/lib/python3.9/dist-packages/pygaggle/model/encode.py in 6 import torch.nn as nn 7 ----> 8 from .tokenize import BatchTokenizer 9 from pygaggle.rerank.base import TextType 10
/usr/local/lib/python3.9/dist-packages/pygaggle/model/tokenize.py in 170 171 --> 172 class SpacySenticizer: 173 nlp = English() 174 nlp.add_pipe(nlp.create_pipe('sentencizer'))
/usr/local/lib/python3.9/dist-packages/pygaggle/model/tokenize.py in SpacySenticizer() 172 class SpacySenticizer: 173 nlp = English() --> 174 nlp.add_pipe(nlp.create_pipe('sentencizer')) 175 176 def init(self, max_paragraph_length: int = None):
/usr/local/lib/python3.9/dist-packages/spacy/language.py in create_pipe(self, name, config) 300 raise KeyError(Errors.E002.format(name=name)) 301 factory = self.factories[name] --> 302 return factory(self, **config) 303 304 def add_pipe(
/usr/local/lib/python3.9/dist-packages/spacy/language.py in factory(nlp, cfg) 1043 def factory(nlp, cfg): 1044 if hasattr(obj, "from_nlp"): -> 1045 return obj.from_nlp(nlp, **cfg) 1046 elif isinstance(obj, class_types): 1047 return obj()
pipes.pyx in spacy.pipeline.pipes.Sentencizer.from_nlp()
TypeError: type() takes 1 or 3 arguments
See pygaggle/issues/327.
jayralencar
commented
1 year ago jayralencar
commented
1 year ago
I am running the notebook on Colab on Python v3.9.16 Kindly help me resolve the issue @rodrigonogueira4, @jayralencar.
1 from pygaggle.rerank.base import Query, Text ----> 2 from pygaggle.rerank.transformer import MonoT5 3 4 reranker = MonoT5(model=model)
7 frames /usr/local/lib/python3.9/dist-packages/pygaggle/rerank/transformer.py in
11 import torch
12 from .base import Reranker, Query, Text
---> 13 from .similarity import SimilarityMatrixProvider
14 from pygaggle.model import (BatchTokenizer,
15 LongBatchEncoder,
/usr/local/lib/python3.9/dist-packages/pygaggle/rerank/similarity.py in
3 import torch
4
----> 5 from pygaggle.model.encode import SingleEncoderOutput
6
7
/usr/local/lib/python3.9/dist-packages/pygaggle/model/init.py in
4
5 from .decode import
----> 6 from .encode import
7 from .evaluate import
8 from .serialize import
/usr/local/lib/python3.9/dist-packages/pygaggle/model/encode.py in
6 import torch.nn as nn
7
----> 8 from .tokenize import BatchTokenizer
9 from pygaggle.rerank.base import TextType
10
/usr/local/lib/python3.9/dist-packages/pygaggle/model/tokenize.py in
170
171
--> 172 class SpacySenticizer:
173 nlp = English()
174 nlp.add_pipe(nlp.create_pipe('sentencizer'))
/usr/local/lib/python3.9/dist-packages/pygaggle/model/tokenize.py in SpacySenticizer() 172 class SpacySenticizer: 173 nlp = English() --> 174 nlp.add_pipe(nlp.create_pipe('sentencizer')) 175 176 def init(self, max_paragraph_length: int = None):
/usr/local/lib/python3.9/dist-packages/spacy/language.py in create_pipe(self, name, config) 300 raise KeyError(Errors.E002.format(name=name)) 301 factory = self.factories[name] --> 302 return factory(self, **config) 303 304 def add_pipe(
/usr/local/lib/python3.9/dist-packages/spacy/language.py in factory(nlp, cfg) 1043 def factory(nlp, cfg): 1044 if hasattr(obj, "from_nlp"): -> 1045 return obj.from_nlp(nlp, **cfg) 1046 elif isinstance(obj, class_types): 1047 return obj()
pipes.pyx in spacy.pipeline.pipes.Sentencizer.from_nlp()
TypeError: type() takes 1 or 3 arguments