fix task to model; handle exception

modelscope/hub/utils/utils.py

@@ -92,9 +92,12 @@ def file_integrity_validation(file_path, expected_sha256):
 def create_library_statistics(method: str,
                               name: str,
                               cn_name: Optional[str]):
-    path = f'{get_endpoint()}/api/v1/statistics/library'
-    headers = {'user-agent': ModelScopeConfig.get_user_agent()}
-    params = {"Method": method, "Name": name, "CnName": cn_name}
-    r = requests.post(path, params=params, headers=headers)
-    r.raise_for_status()
+    try:
+        path = f'{get_endpoint()}/api/v1/statistics/library'
+        headers = {'user-agent': ModelScopeConfig.get_user_agent()}
+        params = {"Method": method, "Name": name, "CnName": cn_name}
+        r = requests.post(path, params=params, headers=headers)
+        r.raise_for_status()
+    except Exception:
+        pass
     return

modelscope/pipelines/base.py

@@ -152,7 +152,7 @@ class Pipeline(ABC):
                  **kwargs) -> Union[Dict[str, Any], Generator]:
         # model provider should leave it as it is
         # modelscope library developer will handle this function
-        model_name = self.cfg.task
+        model_name = self.cfg.model.type
         create_library_statistics("pipeline", model_name, None)
         # place model to cpu or gpu
         if (self.model or (self.has_multiple_models and self.models[0])):

modelscope/trainers/trainer.py

@@ -437,7 +437,7 @@ class EpochBasedTrainer(BaseTrainer):
 
     def train(self, checkpoint_path=None, *args, **kwargs):
         self._mode = ModeKeys.TRAIN
-        model_name = self.cfg.task
+        model_name = self.cfg.model.type
         create_library_statistics("train", model_name, None)
 
         if self.train_dataset is None:
@@ -459,7 +459,7 @@ class EpochBasedTrainer(BaseTrainer):
         self.train_loop(self.train_dataloader)
 
     def evaluate(self, checkpoint_path=None):
-        model_name = self.cfg.task
+        model_name = self.cfg.model.type
         create_library_statistics("evaluate", model_name, None)
         if checkpoint_path is not None and os.path.isfile(checkpoint_path):
             from modelscope.trainers.hooks import CheckpointHook