fix

modelscope · Jul 8, 2024 · be9253a · be9253a
1 parent 7562138
commit be9253a
Show file tree

Hide file tree

Showing 2 changed files with 5 additions and 3 deletions.
diff --git a/swift/llm/utils/model.py b/swift/llm/utils/model.py
@@ -5271,12 +5271,13 @@ def get_model_tokenizer_llava(model_dir: str,
                               **kwargs):
     llm_model_type = kwargs.pop('llm_model_type')
     if 'local_repo_path' in kwargs:
-        repo_path = kwargs['local_repo_path']
+        local_repo_path = kwargs['local_repo_path']
     elif 'next' in llm_model_type:
         repo_path = 'https://github.com/LLaVA-VL/LLaVA-NeXT'
+        local_repo_path = _git_clone_github(repo_path)
     else:
         repo_path = 'https://github.com/haotian-liu/LLaVA'
-    local_repo_path = _git_clone_github(repo_path)
+        local_repo_path = _git_clone_github(repo_path)
     sys.path.append(os.path.join(local_repo_path))
 
     if llm_model_type == 'mistral':

diff --git a/swift/llm/utils/preprocess.py b/swift/llm/utils/preprocess.py
@@ -24,7 +24,7 @@ def _reduce_columns(cls: type) -> type:
     cls._patching = True
 
     def new_call_func(self, dataset: HfDataset) -> HfDataset:
-        self.column_state = set()
+        self.column_state = set(['images', 'videos', 'audios'])
         dataset = call_func(self, dataset)
         for k in dataset.features.keys():
             if k not in self.column_state:
@@ -240,6 +240,7 @@ def preprocess(self, d: Dict[str, Any]) -> Dict[str, Any]:
                 hr.append([q[self.from_key], r[self.from_key]])
             query = conversations[-2][self.value_key]
             query_role = conversations[-2][self.from_key]
+            query_role = query_role if query_role == 'tool' else 'user'
             response = conversations[-1][self.value_key]
             system = sys
             history = h