fix tensorboard monitor

Author: pan-x-c

trinity/trainer/trainer.py

@@ -12,7 +12,7 @@
 import ray
 
 from trinity.buffer import get_buffer_reader
-from trinity.common.config import Config, TrainerConfig
+from trinity.common.config import Config
 from trinity.common.constants import AlgorithmType
 from trinity.common.experience import Experiences
 from trinity.utils.log import get_logger
@@ -37,7 +37,7 @@ def __init__(self, config: Config) -> None:
             if self.config.trainer.sft_warmup_iteration > 0
             else None
         )
-        self.engine = get_trainer_wrapper(config.trainer)
+        self.engine = get_trainer_wrapper(config)
 
     def prepare(self) -> None:
         """Prepare the trainer."""
@@ -146,9 +146,9 @@ def shutdown(self) -> None:
         """Shutdown the engine."""
 
 
-def get_trainer_wrapper(config: TrainerConfig) -> TrainEngineWrapper:
+def get_trainer_wrapper(config: Config) -> TrainEngineWrapper:
     """Get a trainer wrapper."""
-    if config.trainer_type == "verl":
+    if config.trainer.trainer_type == "verl":
         from trinity.trainer.verl_trainer import VerlPPOTrainerWrapper
 
         return VerlPPOTrainerWrapper(config)

trinity/trainer/verl_trainer.py

@@ -13,7 +13,7 @@
 from verl.utils import hf_tokenizer
 from verl.utils.fs import copy_local_path_from_hdfs
 
-from trinity.common.config import TrainerConfig
+from trinity.common.config import Config
 from trinity.common.constants import AlgorithmType
 from trinity.common.experience import Experiences
 from trinity.trainer.trainer import TrainEngineWrapper
@@ -71,8 +71,9 @@ class VerlPPOTrainerWrapper(RayPPOTrainer, TrainEngineWrapper):
 
     def __init__(
         self,
-        train_config: TrainerConfig,
+        global_config: Config,
     ):
+        train_config = global_config.trainer
         pprint(train_config.trainer_config)
         config = OmegaConf.structured(train_config.trainer_config)
         # download the checkpoint from hdfs
@@ -134,7 +135,7 @@ def __init__(
             project=config.trainer.project_name,
             name=config.trainer.experiment_name,
             role="trainer",
-            config=train_config,
+            config=global_config,
         )
         self.reset_experiences_example_table()