[Trainer] Change num_train_epochs default value (#8113)

* change num_train_epochs default value

* update docs

docs/trainer.md

@@ -321,9 +321,9 @@ Trainer 是一个简单，但功能完整的 Paddle训练和评估模块，并
   --num_train_epochs
                         要执行的训练 epoch 总数（如果不是整数，将在停止训练
                         之前执行最后一个 epoch 的小数部分百分比）。
-                        (`float`, 可选, 默认为 3.0):
+                        (`float`, 可选, 默认为 1.0):
 
-                        Total number of training epochs to perform. (default:3.0)
+                        Total number of training epochs to perform. (default:1.0)
 
   --max_steps
                         如果设置为正数，则表示要执行的训练步骤总数。

paddlenlp/trainer/training_args.py

@@ -127,7 +127,7 @@ class TrainingArguments:
             The epsilon hyperparameter for the [`AdamW`] optimizer.
         max_grad_norm (`float`, *optional*, defaults to 1.0):
             Maximum gradient norm (for gradient clipping).
-        num_train_epochs(`float`, *optional*, defaults to 3.0):
+        num_train_epochs(`float`, *optional*, defaults to 1.0):
             Total number of training epochs to perform (if not an integer, will perform the decimal part percents of
             the last epoch before stopping training).
         max_steps (`int`, *optional*, defaults to -1):
@@ -391,7 +391,7 @@ class TrainingArguments:
     adam_epsilon: float = field(default=1e-8, metadata={"help": "Epsilon for AdamW optimizer."})
     max_grad_norm: float = field(default=1.0, metadata={"help": "Max gradient norm."})
 
-    num_train_epochs: float = field(default=3.0, metadata={"help": "Total number of training epochs to perform."})
+    num_train_epochs: float = field(default=1.0, metadata={"help": "Total number of training epochs to perform."})
     max_steps: int = field(
         default=-1,
         metadata={"help": "If > 0: set total number of training steps to perform. Override num_train_epochs."},