pre-commit lint

modelscope/models/__init__.py

@@ -5,13 +5,7 @@ from .audio.tts.vocoder import Hifigan16k
 from .base import Model
 from .builder import MODELS, build_model
 from .multi_model import OfaForImageCaptioning
-from .nlp import (
-    BertForSequenceClassification,
-    SbertForNLI,
-    SbertForSentenceSimilarity,
-    SbertForSentimentClassification,
-    SbertForZeroShotClassification,
-    StructBertForMaskedLM,
-    VecoForMaskedLM,
-    SbertForTokenClassification,
-)
+from .nlp import (BertForSequenceClassification, SbertForNLI,
+                  SbertForSentenceSimilarity, SbertForSentimentClassification,
+                  SbertForTokenClassification, SbertForZeroShotClassification,
+                  StructBertForMaskedLM, VecoForMaskedLM)

modelscope/models/nlp/__init__.py

@@ -1,10 +1,10 @@
 from .bert_for_sequence_classification import *  # noqa F403
 from .masked_language_model import *  # noqa F403
-from .sbert_for_nli import *  # noqa F403
 from .palm_for_text_generation import *  # noqa F403
+from .sbert_for_nli import *  # noqa F403
 from .sbert_for_sentence_similarity import *  # noqa F403
-from .sbert_for_token_classification import *  # noqa F403
 from .sbert_for_sentiment_classification import *  # noqa F403
+from .sbert_for_token_classification import *  # noqa F403
+from .sbert_for_zero_shot_classification import *  # noqa F403
 from .space.dialog_intent_prediction_model import *  # noqa F403
 from .space.dialog_modeling_model import *  # noqa F403
-from .sbert_for_zero_shot_classification import *  # noqa F403

modelscope/models/nlp/masked_language_model.py

@@ -2,10 +2,10 @@ from typing import Any, Dict, Optional, Union
 
 import numpy as np
 
+from ...metainfo import Models
 from ...utils.constant import Tasks
 from ..base import Model, Tensor
 from ..builder import MODELS
-from ...metainfo import Models
 
 __all__ = ['StructBertForMaskedLM', 'VecoForMaskedLM', 'MaskedLMModelBase']
 
@@ -27,7 +27,7 @@ class MaskedLMModelBase(Model):
 
     @property
     def config(self):
-        if hasattr(self.model, "config"):
+        if hasattr(self.model, 'config'):
             return self.model.config
         return None
 

modelscope/models/nlp/sbert_for_nli.py

@@ -1,7 +1,8 @@
-from ...utils.constant import Tasks
-from .sbert_for_sequence_classification import SbertForSequenceClassificationBase
-from ..builder import MODELS
 from ...metainfo import Models
+from ...utils.constant import Tasks
+from ..builder import MODELS
+from .sbert_for_sequence_classification import \
+    SbertForSequenceClassificationBase
 
 __all__ = ['SbertForNLI']
 
@@ -17,5 +18,6 @@ class SbertForNLI(SbertForSequenceClassificationBase):
             model_cls (Optional[Any], optional): model loader, if None, use the
                 default loader to load model weights, by default None.
         """
-        super().__init__(model_dir, *args, model_args={"num_labels": 3}, **kwargs)
+        super().__init__(
+            model_dir, *args, model_args={'num_labels': 3}, **kwargs)
         assert self.model.config.num_labels == 3

modelscope/models/nlp/sbert_for_sentence_similarity.py

@@ -1,7 +1,8 @@
 from modelscope.metainfo import Models
 from modelscope.utils.constant import Tasks
-from .sbert_for_sequence_classification import SbertForSequenceClassificationBase
 from ..builder import MODELS
+from .sbert_for_sequence_classification import \
+    SbertForSequenceClassificationBase
 
 __all__ = ['SbertForSentenceSimilarity']
 
@@ -18,6 +19,7 @@ class SbertForSentenceSimilarity(SbertForSequenceClassificationBase):
             model_cls (Optional[Any], optional): model loader, if None, use the
                 default loader to load model weights, by default None.
         """
-        super().__init__(model_dir, *args, model_args={"num_labels": 2}, **kwargs)
+        super().__init__(
+            model_dir, *args, model_args={'num_labels': 2}, **kwargs)
         self.model_dir = model_dir
         assert self.model.config.num_labels == 2

modelscope/models/nlp/sbert_for_sentiment_classification.py

@@ -1,14 +1,14 @@
-from modelscope.utils.constant import Tasks
-from .sbert_for_sequence_classification import SbertForSequenceClassificationBase
-from ..builder import MODELS
 from modelscope.metainfo import Models
+from modelscope.utils.constant import Tasks
+from ..builder import MODELS
+from .sbert_for_sequence_classification import \
+    SbertForSequenceClassificationBase
 
 __all__ = ['SbertForSentimentClassification']
 
 
 @MODELS.register_module(
-    Tasks.sentiment_classification,
-    module_name=Models.structbert)
+    Tasks.sentiment_classification, module_name=Models.structbert)
 class SbertForSentimentClassification(SbertForSequenceClassificationBase):
 
     def __init__(self, model_dir: str, *args, **kwargs):
@@ -19,5 +19,6 @@ class SbertForSentimentClassification(SbertForSequenceClassificationBase):
             model_cls (Optional[Any], optional): model loader, if None, use the
                 default loader to load model weights, by default None.
         """
-        super().__init__(model_dir, *args, model_args={"num_labels": 2}, **kwargs)
+        super().__init__(
+            model_dir, *args, model_args={'num_labels': 2}, **kwargs)
         assert self.model.config.num_labels == 2

modelscope/models/nlp/sbert_for_sequence_classification.py

@@ -1,11 +1,13 @@
-from torch import nn
-from typing import Any, Dict
-from ..base import Model
-import numpy as np
-import json
 import os
+from typing import Any, Dict
+
+import json
+import numpy as np
 import torch
-from sofa.models.sbert.modeling_sbert import SbertPreTrainedModel, SbertModel
+from sofa.models.sbert.modeling_sbert import SbertModel, SbertPreTrainedModel
+from torch import nn
+
+from ..base import Model
 
 
 class SbertTextClassfier(SbertPreTrainedModel):
@@ -27,9 +29,7 @@ class SbertTextClassfier(SbertPreTrainedModel):
         pooled_output = outputs[1]
         pooled_output = self.dropout(pooled_output)
         logits = self.classifier(pooled_output)
-        return {
-            "logits": logits
-        }
+        return {'logits': logits}
 
 
 class SbertForSequenceClassificationBase(Model):
@@ -38,13 +38,17 @@ class SbertForSequenceClassificationBase(Model):
         super().__init__(model_dir, *args, **kwargs)
         if model_args is None:
             model_args = {}
-        self.model = SbertTextClassfier.from_pretrained(model_dir, **model_args)
+        self.model = SbertTextClassfier.from_pretrained(
+            model_dir, **model_args)
         self.id2label = {}
         self.label_path = os.path.join(self.model_dir, 'label_mapping.json')
         if os.path.exists(self.label_path):
             with open(self.label_path) as f:
                 self.label_mapping = json.load(f)
-            self.id2label = {idx: name for name, idx in self.label_mapping.items()}
+            self.id2label = {
+                idx: name
+                for name, idx in self.label_mapping.items()
+            }
 
     def train(self):
         return self.model.train()
@@ -59,7 +63,7 @@ class SbertForSequenceClassificationBase(Model):
         return self.model.forward(input_ids, token_type_ids)
 
     def postprocess(self, input, **kwargs):
-        logits = input["logits"]
+        logits = input['logits']
         probs = logits.softmax(-1).numpy()
         pred = logits.argmax(-1).numpy()
         logits = logits.numpy()

modelscope/models/nlp/sbert_for_token_classification.py

@@ -34,7 +34,7 @@ class SbertForTokenClassification(Model):
 
     def eval(self):
         return self.model.eval()
-    
+
     def forward(self, input: Dict[str,
                                   Any]) -> Dict[str, Union[str, np.ndarray]]:
         """return the result by the model
@@ -54,8 +54,9 @@ class SbertForTokenClassification(Model):
         input_ids = torch.tensor(input['input_ids']).unsqueeze(0)
         return {**self.model(input_ids), 'text': input['text']}
 
-    def postprocess(self, input: Dict[str, Tensor], **kwargs) -> Dict[str, Tensor]:
-        logits = input["logits"]
+    def postprocess(self, input: Dict[str, Tensor],
+                    **kwargs) -> Dict[str, Tensor]:
+        logits = input['logits']
         pred = torch.argmax(logits[0], dim=-1)
         pred = pred.numpy()
         rst = {'predictions': pred, 'logits': logits, 'text': input['text']}

modelscope/models/nlp/sbert_for_zero_shot_classification.py

@@ -3,16 +3,15 @@ from typing import Any, Dict
 import numpy as np
 
 from modelscope.utils.constant import Tasks
+from ...metainfo import Models
 from ..base import Model
 from ..builder import MODELS
-from ...metainfo import Models
 
 __all__ = ['SbertForZeroShotClassification']
 
 
 @MODELS.register_module(
-    Tasks.zero_shot_classification,
-    module_name=Models.structbert)
+    Tasks.zero_shot_classification, module_name=Models.structbert)
 class SbertForZeroShotClassification(Model):
 
     def __init__(self, model_dir: str, *args, **kwargs):
@@ -31,7 +30,7 @@ class SbertForZeroShotClassification(Model):
 
     def eval(self):
         return self.model.eval()
-    
+
     def forward(self, input: Dict[str, Any]) -> Dict[str, np.ndarray]:
         """return the result by the model
 

modelscope/models/nlp/space/model/gen_unified_transformer.py

@@ -3,8 +3,7 @@ IntentUnifiedTransformer
 """
 import torch
 
-from .unified_transformer import \
-    UnifiedTransformer
+from .unified_transformer import UnifiedTransformer
 
 
 class GenUnifiedTransformer(UnifiedTransformer):

modelscope/models/nlp/space/model/unified_transformer.py

@@ -7,10 +7,9 @@ import torch
 import torch.nn as nn
 import torch.nn.functional as F
 
-from .model_base import ModelBase
 from ..modules.embedder import Embedder
-from ..modules.transformer_block import \
-    TransformerBlock
+from ..modules.transformer_block import TransformerBlock
+from .model_base import ModelBase
 
 
 class UnifiedTransformer(ModelBase):

modelscope/models/nlp/space/modules/transformer_block.py

@@ -6,8 +6,7 @@ import torch
 import torch.nn as nn
 
 from .feedforward import FeedForward
-from .multihead_attention import \
-    MultiheadAttention
+from .multihead_attention import MultiheadAttention
 
 
 class TransformerBlock(nn.Module):

modelscope/pipelines/nlp/fill_mask_pipeline.py

@@ -1,15 +1,14 @@
-from typing import Dict, Optional, Union, Any
+from typing import Any, Dict, Optional, Union
 
 import torch
 
+from ...metainfo import Pipelines
 from ...models import Model
-from ...models.nlp.masked_language_model import \
-    MaskedLMModelBase
+from ...models.nlp.masked_language_model import MaskedLMModelBase
 from ...preprocessors import FillMaskPreprocessor
 from ...utils.constant import Tasks
 from ..base import Pipeline, Tensor
 from ..builder import PIPELINES
-from ...metainfo import Pipelines
 
 __all__ = ['FillMaskPipeline']
 
@@ -20,7 +19,7 @@ class FillMaskPipeline(Pipeline):
     def __init__(self,
                  model: Union[MaskedLMModelBase, str],
                  preprocessor: Optional[FillMaskPreprocessor] = None,
-                 first_sequence="sentense",
+                 first_sequence='sentense',
                  **kwargs):
         """use `model` and `preprocessor` to create a nlp fill mask pipeline for prediction
 
@@ -38,7 +37,8 @@ class FillMaskPipeline(Pipeline):
                 first_sequence=first_sequence,
                 second_sequence=None)
         fill_mask_model.eval()
-        super().__init__(model=fill_mask_model, preprocessor=preprocessor, **kwargs)
+        super().__init__(
+            model=fill_mask_model, preprocessor=preprocessor, **kwargs)
         self.preprocessor = preprocessor
         self.tokenizer = preprocessor.tokenizer
         self.mask_id = {'veco': 250001, 'sbert': 103}

modelscope/pipelines/nlp/nli_pipeline.py

@@ -1,31 +1,28 @@
 import uuid
 from typing import Any, Dict, Union
-import torch
-import uuid
-from typing import Any, Dict, Union
 
 import numpy as np
+import torch
 
-from ..base import Pipeline
-from ..builder import PIPELINES
 from ...metainfo import Pipelines
 from ...models import Model
 from ...models.nlp import SbertForNLI
 from ...preprocessors import NLIPreprocessor
 from ...utils.constant import Tasks
+from ..base import Pipeline
+from ..builder import PIPELINES
 
 __all__ = ['NLIPipeline']
 
 
-@PIPELINES.register_module(
-    Tasks.nli, module_name=Pipelines.nli)
+@PIPELINES.register_module(Tasks.nli, module_name=Pipelines.nli)
 class NLIPipeline(Pipeline):
 
     def __init__(self,
                  model: Union[SbertForNLI, str],
                  preprocessor: NLIPreprocessor = None,
-                 first_sequence="first_sequence",
-                 second_sequence="second_sequence",
+                 first_sequence='first_sequence',
+                 second_sequence='second_sequence',
                  **kwargs):
         """use `model` and `preprocessor` to create a nlp text classification pipeline for prediction
 
@@ -51,7 +48,8 @@ class NLIPipeline(Pipeline):
         with torch.no_grad():
             return super().forward(inputs, **forward_params)
 
-    def postprocess(self, inputs: Dict[str, Any], **postprocess_params) -> Dict[str, str]:
+    def postprocess(self, inputs: Dict[str, Any],
+                    **postprocess_params) -> Dict[str, str]:
         """process the prediction results
 
         Args:

modelscope/pipelines/nlp/sentence_similarity_pipeline.py

@@ -2,11 +2,12 @@ from typing import Any, Dict, Union
 
 import numpy as np
 import torch
+
 from ...metainfo import Pipelines
+from ...models import Model
 from ...models.nlp import SbertForSentenceSimilarity
 from ...preprocessors import SequenceClassificationPreprocessor
 from ...utils.constant import Tasks
-from ...models import Model
 from ..base import Input, Pipeline
 from ..builder import PIPELINES
 
@@ -20,8 +21,8 @@ class SentenceSimilarityPipeline(Pipeline):
     def __init__(self,
                  model: Union[Model, str],
                  preprocessor: SequenceClassificationPreprocessor = None,
-                 first_sequence="first_sequence",
-                 second_sequence="second_sequence",
+                 first_sequence='first_sequence',
+                 second_sequence='second_sequence',
                  **kwargs):
         """use `model` and `preprocessor` to create a nlp sentence similarity pipeline for prediction
 
@@ -50,7 +51,8 @@ class SentenceSimilarityPipeline(Pipeline):
         with torch.no_grad():
             return super().forward(inputs, **forward_params)
 
-    def postprocess(self, inputs: Dict[str, Any], **postprocess_params) -> Dict[str, str]:
+    def postprocess(self, inputs: Dict[str, Any],
+                    **postprocess_params) -> Dict[str, str]:
         """process the prediction results
 
         Args:

modelscope/pipelines/nlp/sentiment_classification_pipeline.py

@@ -1,17 +1,18 @@
 import os
 import uuid
 from typing import Any, Dict, Union
-import torch
+
 import json
 import numpy as np
+import torch
 
+from ...metainfo import Pipelines
+from ...models import Model
 from ...models.nlp import SbertForSentimentClassification
 from ...preprocessors import SentimentClassificationPreprocessor
 from ...utils.constant import Tasks
-from ...models import Model
 from ..base import Input, Pipeline
 from ..builder import PIPELINES
-from ...metainfo import Pipelines
 
 __all__ = ['SentimentClassificationPipeline']
 
@@ -24,8 +25,8 @@ class SentimentClassificationPipeline(Pipeline):
     def __init__(self,
                  model: Union[SbertForSentimentClassification, str],
                  preprocessor: SentimentClassificationPreprocessor = None,
-                 first_sequence="first_sequence",
-                 second_sequence="second_sequence",
+                 first_sequence='first_sequence',
+                 second_sequence='second_sequence',
                  **kwargs):
         """use `model` and `preprocessor` to create a nlp text classification pipeline for prediction
 
@@ -52,7 +53,8 @@ class SentimentClassificationPipeline(Pipeline):
         with torch.no_grad():
             return super().forward(inputs, **forward_params)
 
-    def postprocess(self, inputs: Dict[str, Any], **postprocess_params) -> Dict[str, str]:
+    def postprocess(self, inputs: Dict[str, Any],
+                    **postprocess_params) -> Dict[str, str]:
         """process the prediction results
 
         Args:

modelscope/pipelines/nlp/text_generation_pipeline.py

@@ -1,5 +1,7 @@
-from typing import Dict, Optional, Union, Any
+from typing import Any, Dict, Optional, Union
+
 import torch
+
 from ...metainfo import Pipelines
 from ...models import Model
 from ...models.nlp import PalmForTextGeneration
@@ -42,7 +44,8 @@ class TextGenerationPipeline(Pipeline):
         with torch.no_grad():
             return super().forward(inputs, **forward_params)
 
-    def postprocess(self, inputs: Dict[str, Tensor], **postprocess_params) -> Dict[str, str]:
+    def postprocess(self, inputs: Dict[str, Tensor],
+                    **postprocess_params) -> Dict[str, str]:
         """process the prediction results
 
         Args:

modelscope/pipelines/nlp/word_segmentation_pipeline.py

@@ -1,5 +1,7 @@
 from typing import Any, Dict, Optional, Union
+
 import torch
+
 from ...metainfo import Pipelines
 from ...models import Model
 from ...models.nlp import SbertForTokenClassification
@@ -42,7 +44,8 @@ class WordSegmentationPipeline(Pipeline):
         with torch.no_grad():
             return super().forward(inputs, **forward_params)
 
-    def postprocess(self, inputs: Dict[str, Any], **postprocess_params) -> Dict[str, str]:
+    def postprocess(self, inputs: Dict[str, Any],
+                    **postprocess_params) -> Dict[str, str]:
         """process the prediction results
 
         Args:

modelscope/pipelines/nlp/zero_shot_classification_pipeline.py

@@ -1,16 +1,17 @@
 import os
 import uuid
 from typing import Any, Dict, Union
-import torch
+
 import json
 import numpy as np
+import torch
 from scipy.special import softmax
 
+from ...metainfo import Pipelines
+from ...models import Model
 from ...models.nlp import SbertForZeroShotClassification
 from ...preprocessors import ZeroShotClassificationPreprocessor
 from ...utils.constant import Tasks
-from ...models import Model
-from ...metainfo import Pipelines
 from ..base import Input, Pipeline
 from ..builder import PIPELINES
 

modelscope/preprocessors/nlp.py

@@ -5,8 +5,7 @@ from typing import Any, Dict, Union
 
 from transformers import AutoTokenizer
 
-from ..metainfo import Preprocessors
-from ..metainfo import Models
+from ..metainfo import Models, Preprocessors
 from ..utils.constant import Fields, InputFields
 from ..utils.type_assert import type_assert
 from .base import Preprocessor

modelscope/preprocessors/space/dialog_intent_prediction_preprocessor.py

@@ -3,13 +3,12 @@
 import os
 from typing import Any, Dict
 
-from .fields.intent_field import \
-    IntentBPETextField
 from ...utils.config import Config
 from ...utils.constant import Fields
 from ...utils.type_assert import type_assert
 from ..base import Preprocessor
 from ..builder import PREPROCESSORS
+from .fields.intent_field import IntentBPETextField
 
 __all__ = ['DialogIntentPredictionPreprocessor']
 

modelscope/preprocessors/space/dialog_modeling_preprocessor.py

@@ -3,13 +3,12 @@
 import os
 from typing import Any, Dict
 
-from .fields.gen_field import \
-    MultiWOZBPETextField
-from ..base import Preprocessor
-from ..builder import PREPROCESSORS
 from ...utils.config import Config
 from ...utils.constant import Fields
 from ...utils.type_assert import type_assert
+from ..base import Preprocessor
+from ..builder import PREPROCESSORS
+from .fields.gen_field import MultiWOZBPETextField
 
 __all__ = ['DialogModelingPreprocessor']
 

modelscope/preprocessors/space/fields/gen_field.py

@@ -8,10 +8,10 @@ from itertools import chain
 
 import numpy as np
 
-from ..tokenizer import Tokenizer
 from ....utils.nlp.space import ontology, utils
 from ....utils.nlp.space.db_ops import MultiWozDB
 from ....utils.nlp.space.utils import list2np
+from ..tokenizer import Tokenizer
 
 
 class BPETextField(object):

modelscope/preprocessors/space/fields/intent_field.py

@@ -14,10 +14,10 @@ import json
 import numpy as np
 from tqdm import tqdm
 
-from ..tokenizer import Tokenizer
 from ....utils.nlp.space import ontology, utils
 from ....utils.nlp.space.scores import hierarchical_set_score
 from ....utils.nlp.space.utils import list2np
+from ..tokenizer import Tokenizer
 
 
 class BPETextField(object):

modelscope/trainers/nlp/space/trainers/intent_trainer.py

@@ -14,8 +14,7 @@ import torch
 from tqdm import tqdm
 from transformers.optimization import AdamW, get_linear_schedule_with_warmup
 
-from ..metrics.metrics_tracker import \
-    MetricsTracker
+from ..metrics.metrics_tracker import MetricsTracker
 
 
 def get_logger(log_path, name='default'):

tests/pipelines/test_word_segmentation.py

@@ -19,8 +19,7 @@ class WordSegmentationTest(unittest.TestCase):
     def test_run_by_direct_model_download(self):
         cache_path = snapshot_download(self.model_id)
         tokenizer = TokenClassifcationPreprocessor(cache_path)
-        model = SbertForTokenClassification(
-            cache_path, tokenizer=tokenizer)
+        model = SbertForTokenClassification(cache_path, tokenizer=tokenizer)
         pipeline1 = WordSegmentationPipeline(model, preprocessor=tokenizer)
         pipeline2 = pipeline(
             Tasks.word_segmentation, model=model, preprocessor=tokenizer)