diff --git a/examples/opensora_pku/examples/rec_video.py b/examples/opensora_pku/examples/rec_video.py
index f905627c00..5c3a4d1f6f 100644
--- a/examples/opensora_pku/examples/rec_video.py
+++ b/examples/opensora_pku/examples/rec_video.py
@@ -134,7 +134,7 @@ def main(args):
             [k.replace("network.", "") if k.startswith("network.") else k, v] for k, v in state_dict.items()
         )
     else:
-        None
+        state_dict = None
     kwarg = {"state_dict": state_dict, "use_safetensors": True}
     vae = CausalVAEModelWrapper(args.ae_path, **kwarg)
 
diff --git a/examples/opensora_pku/examples/rec_video_folder.py b/examples/opensora_pku/examples/rec_video_folder.py
index 74795799dd..20426592c7 100644
--- a/examples/opensora_pku/examples/rec_video_folder.py
+++ b/examples/opensora_pku/examples/rec_video_folder.py
@@ -71,7 +71,7 @@ def main(args):
             [k.replace("network.", "") if k.startswith("network.") else k, v] for k, v in state_dict.items()
         )
     else:
-        None
+        state_dict = None
     kwarg = {"state_dict": state_dict, "use_safetensors": True}
     vae = CausalVAEModelWrapper(args.ae_path, **kwarg)
 
diff --git a/examples/opensora_pku/opensora/train/train_t2v_diffusers.py b/examples/opensora_pku/opensora/train/train_t2v_diffusers.py
index 02391ab4d1..3c02305d96 100644
--- a/examples/opensora_pku/opensora/train/train_t2v_diffusers.py
+++ b/examples/opensora_pku/opensora/train/train_t2v_diffusers.py
@@ -314,7 +314,8 @@ def main(args):
 
     val_dataloader = None
     if args.validate:
-        assert args.val_data is not None and os.path.exists(args.val_data), "validation dataset must be specified!"
+        assert args.val_data is not None, f"validation dataset must be specified, but got {args.val_data}"
+        assert os.path.exists(args.val_data), f"validation dataset file must exist, but got {args.val_data}"
         print_banner("Validation dataset Loading...")
         val_dataset = getdataset(args, dataset_file=args.val_data)
         sampler = (
@@ -406,12 +407,12 @@ def main(args):
 
     if steps_per_sink == dataloader_size:
         logger.info(
-            f"Number of training steps: {total_train_steps}; Number of epochs: {args.num_train_epochs}; "
+            f"Number of training steps: {total_train_steps}, Number of epochs: {args.num_train_epochs}, "
             f"Number of batches in a epoch (dataloader size): {dataloader_size}"
         )
     else:
         logger.info(
-            f"Number of training steps: {total_train_steps}; Number of sink epochs: {sink_epochs}; Number of batches in a sink (sink_size): {steps_per_sink}"
+            f"Number of training steps: {total_train_steps}, Number of sink epochs: {sink_epochs}, Number of batches in a sink (sink_size): {steps_per_sink}"
         )
     if args.checkpointing_steps is None:
         ckpt_save_interval = args.ckpt_save_interval
@@ -425,7 +426,7 @@ def main(args):
             ckpt_save_interval = max(1, args.checkpointing_steps // steps_per_sink)
             if args.checkpointing_steps % steps_per_sink != 0:
                 logger.warning(
-                    f"`checkpointing_steps` must be times of sink size or dataset_size under dataset sink mode."
+                    "`checkpointing_steps` must be times of sink size or dataset_size under dataset sink mode."
                     f"Checkpoint will be saved every {ckpt_save_interval * steps_per_sink} steps."
                 )
     if step_mode != args.step_mode:
@@ -610,13 +611,16 @@ def main(args):
             integrated_save=integrated_save,
             save_training_resume=save_training_resume,
         )
-        rec_cb = PerfRecorderCallback(
-            save_dir=args.output_dir,
-            file_name="result_val.log",
-            resume=args.resume_from_checkpoint,
-            metric_names=list(metrics.keys()),
-        )
-        callback.extend([save_cb, rec_cb])
+        callback.append(save_cb)
+        if args.validate:
+            assert metrics is not None, "Val during training must set the metric functions"
+            rec_cb = PerfRecorderCallback(
+                save_dir=args.output_dir,
+                file_name="result_val.log",
+                resume=args.resume_from_checkpoint,
+                metric_names=list(metrics.keys()),
+            )
+            callback.append(rec_cb)
         if args.profile:
             callback.append(ProfilerCallbackEpoch(2, 2, "./profile_data"))
     # Train!
@@ -637,10 +641,15 @@ def main(args):
                 f"MindSpore mode[GRAPH(0)/PYNATIVE(1)]: {args.mode}",
                 f"Jit level: {args.jit_level}",
                 f"Distributed mode: {args.use_parallel}"
-                + (f"\nParallel mode: {args.parallel_mode}" if args.use_parallel else "")
+                + (
+                    f"\nParallel mode: {args.parallel_mode}"
+                    + (f"{args.zero_stage}" if args.parallel_mode == "zero" else "")
+                    if args.use_parallel
+                    else ""
+                )
                 + (f"\nsp_size: {args.sp_size}" if args.sp_size != 1 else ""),
-                f"Num params: {num_params:,} (transformer: {num_params_transformer:,}, vae: {num_params_vae:,})",
-                f"Num trainable params: {num_params_trainable:,}",
+                f"Num params: {num_params: , } (transformer: {num_params_transformer: , }, vae: {num_params_vae: , })",
+                f"Num trainable params: {num_params_trainable: , }",
                 f"Transformer model dtype: {model_dtype}",
                 f"Transformer AMP level: {args.amp_level}" if not args.global_bf16 else "Global BF16: True",
                 f"VAE dtype: {vae_dtype} (amp level O2)"