fix bugs

modelscope/models/nlp/plug/plug_for_text_generation.py

@@ -8,12 +8,18 @@ from modelscope.models.nlp.structbert import SbertTokenizer
 from modelscope.models.nlp.utils.distributed import DistributedTorchModel
 from . import DistributedPlug
 from ...base import Tensor
+from modelscope.utils.hub import read_config
 
 __all__ = ['PlugForTextGeneration']
 
 
 @MODELS.register_module(Tasks.text_generation, module_name=Models.plug)
 class PlugForTextGeneration(DistributedTorchModel):
+   
+    def __init__(self, model_dir, cls_token_id, **kwargs):
+        assert cls_token_id is not None
+        super().__init__(model_dir, **kwargs)
+        self.cls_token_id = cls_token_id
 
     def _forward_one(self, input: Dict[str, Any]) -> Dict[str, Tensor]:
         return self.model(**input)
@@ -25,10 +31,9 @@ class PlugForTextGeneration(DistributedTorchModel):
         res = self.model_pool.map(DistributedPlug.forward, [input]*self.world_size)
         return res[0]
 
-    def _instantiate_one(self, model_dir, rank):
-        tokenizer = SbertTokenizer.from_pretrained(model_dir)
-        self.cls_token_id = tokenizer.cls_token_id
-        self.model = DistributedPlug.instantiate(model_dir, rank)
+    def _instantiate_one(self, rank, model_dir):
+        cfg = read_config(model_dir)
+        self.model = DistributedPlug(model_dir, rank, **cfg.model)
 
         
 

modelscope/models/nlp/utils/distributed.py

@@ -15,19 +15,24 @@ class DistributedTorchModel(Model):
         super().__init__(model_dir, *args, **kwargs)
         self.model_pool = None
         self.world_size = None
+    
+    def __getstate__(self):
+        self_dict = self.__dict__.copy()
+        del self_dict['model_pool']
+        return self_dict
 
     @classmethod
-    def _instantiate(cls, model_dir):
-        model = DistributedTorchModel(model_dir=model_dir)
+    def _instantiate(cls, model_dir, **kwargs):
+        model = cls(model_dir=model_dir, **kwargs)
         torch.multiprocessing.set_start_method("spawn")
         cfg = read_config(model_dir)
-        model.world_size = cfg.model.word_size
+        model.world_size = cfg.model.world_size
         ranks = list(range(model.world_size))
         model.model_pool = Pool(model.world_size)
         model.model_pool.map(partial(model._instantiate_one, model_dir=model_dir), ranks)
         return model
 
-    def _instantiate_one(self, model_dir, rank):
+    def _instantiate_one(self, rank, model_dir):
         pass
 
     def forward(self, input: Dict[str, Any]) -> Dict[str, Any]:

modelscope/pipelines/nlp/text_generation_pipeline.py

@@ -55,7 +55,7 @@ class TextGenerationPipeline(Pipeline):
                 first_sequence=first_sequence,
                 second_sequence=None,
                 sequence_length=kwargs.pop('sequence_length', 128))
-        model.eval()
+        # model.eval()
         super().__init__(model=model, preprocessor=preprocessor, **kwargs)
         self.tokenizer = preprocessor.tokenizer
 

modelscope/utils/nlp/fp16/fp16util.py

@@ -18,8 +18,7 @@ import torch.nn as nn
 from torch.autograd import Variable
 from torch._utils import _flatten_dense_tensors, _unflatten_dense_tensors
 
-from sofa.utils import mpu
-
+from modelscope.utils.nlp import mpu
 
 class tofp16(nn.Module):
     """

modelscope/utils/torch_utils.py

@@ -48,10 +48,6 @@ def init_dist(launcher: str, backend: str = 'nccl', **kwargs) -> None:
 
 
 def _init_dist_pytorch(backend: str, **kwargs) -> None:
-    # rank = int(os.environ['RANK'])
-    local_rank = int(os.environ['LOCAL_RANK'])
-
-    #torch.cuda.set_device(local_rank)
     dist.init_process_group(backend=backend, **kwargs)
 
 

tests/pipelines/test_text_generation.py

@@ -9,6 +9,7 @@ from modelscope.pipelines.nlp import TextGenerationPipeline
 from modelscope.preprocessors import TextGenerationPreprocessor
 from modelscope.utils.constant import Tasks
 from modelscope.utils.test_utils import test_level
+from transformers import BertTokenizer
 
 
 class TextGenerationTest(unittest.TestCase):
@@ -38,11 +39,12 @@ class TextGenerationTest(unittest.TestCase):
         self.plug_input = '段誉轻挥折扇，摇了摇头，说'
 
     def test_plug(self):
-        cache_path = "/home/suluyan.sly/model/plug_model"
-        model = PlugForTextGeneration(cache_path)
+        cache_path = "/home/yuze.zyz/MaaS-lib/plug_model"
+        tokenizer = BertTokenizer.from_pretrained(cache_path)
+        model = PlugForTextGeneration._instantiate(cache_path, cls_token_id=tokenizer.cls_token_id)
         preprocessor = TextGenerationPreprocessor(
             cache_path,
-            model.tokenizer,
+            tokenizer,
             first_sequence='sentence',
             second_sequence=None)
         pipeline1 = TextGenerationPipeline(model, preprocessor)