[Bump] Bump version to 0.3.7 (#1733)

opencompass/__init__.py

@@ -1,8 +1,9 @@
-__version__ = '0.3.6'
+__version__ = '0.3.7'
 
 
 def _warn_about_config_migration():
     import warnings
+
     warnings.warn(
         'Starting from v0.4.0, all AMOTIC configuration files currently '
         'located in `./configs/datasets`, `./configs/models`, and '
@@ -10,7 +11,8 @@ def _warn_about_config_migration():
         '`opencompass/configs/` package. Please update your configuration '
         'file paths accordingly.',
         UserWarning,  # Changed to UserWarning
-        stacklevel=2)
+        stacklevel=2,
+    )
 
 
 # Trigger the warning

opencompass/configs/models/deepseek/lmdeploy_deepseek_v2_lite.py

@@ -3,7 +3,7 @@ from opencompass.models import TurboMindModelwithChatTemplate
 models = [
     dict(
         type=TurboMindModelwithChatTemplate,
-        abbr='deepseek-v2_lite-turbomind',
+        abbr='deepseek-v2_lite-chat-turbomind',
         path='deepseek-ai/DeepSeek-V2-Lite-Chat',
         engine_config=dict(
             session_len=7168,

opencompass/configs/models/gemma/lmdeploy_gemma_27b.py

@@ -0,0 +1,17 @@
+from opencompass.models import TurboMindModel
+
+models = [
+    dict(
+        type=TurboMindModel,
+        abbr='gemma-2-27b-turbomind',
+        path='google/gemma-2-27b',
+        engine_config=dict(session_len=16384, max_batch_size=16, tp=2),
+        gen_config=dict(
+            top_k=1, temperature=1e-6, top_p=0.9, max_new_tokens=4096
+        ),
+        max_seq_len=16384,
+        max_out_len=4096,
+        batch_size=16,
+        run_cfg=dict(num_gpus=2),
+    )
+]

opencompass/configs/models/gemma/lmdeploy_gemma_9b.py

@@ -0,0 +1,17 @@
+from opencompass.models import TurboMindModel
+
+models = [
+    dict(
+        type=TurboMindModel,
+        abbr='gemma-2-9b-turbomind',
+        path='google/gemma-2-9b',
+        engine_config=dict(session_len=16384, max_batch_size=16, tp=1),
+        gen_config=dict(
+            top_k=1, temperature=1e-6, top_p=0.9, max_new_tokens=4096
+        ),
+        max_seq_len=16384,
+        max_out_len=4096,
+        batch_size=16,
+        run_cfg=dict(num_gpus=1),
+    )
+]