modify default xpu and gpu to false (#5074)

PaddleNLP/legacy/pretrain_language_models/BERT/README.md

@@ -202,7 +202,6 @@ DATA_PATH=/path/to/xnli/data/
 CKPT_PATH=/path/to/save/checkpoints/
 
 python -u run_classifier.py --task_name ${TASK_NAME} \
-                   --use_cuda false \
                    --use_xpu true \
                    --do_train true \
                    --do_val true \
@@ -232,6 +231,13 @@ python -u run_classifier.py --task_name ${TASK_NAME} \
 [dev evaluation] ave loss: 0.622958, ave acc: 0.770281, elapsed time: 8.946956 s
 ```
 
+xpu训练结束后，验证集上的测试结果：
+
+```
+[dev evaluation] ave loss: 0.620479, ave acc: 0.762249, elapsed time: 70.831693 s
+[test evaluation] ave loss: 0.616955, ave acc: 0.762275, elapsed time: 142.251840 s
+```
+
 ### 阅读理解 SQuAD
 
 下载 SQuAD 的数据以及测评脚本到 `$SQUAD_APTH` 目录。

PaddleNLP/legacy/pretrain_language_models/BERT/run_classifier.py

@@ -100,8 +100,8 @@ run_type_g.add_arg("profiler_path",                str,    './', "the profiler o
 run_type_g.add_arg("is_profiler",                  int,    0,     "the profiler switch. (used for benchmark)")
 run_type_g.add_arg("max_iter",                     int,    0,     "the max batch nums to train. (used for benchmark)")
 
-run_type_g.add_arg("use_cuda",                     bool,   True,  "If set, use GPU for training.")
-run_type_g.add_arg("use_xpu",                      bool,   True,  "If set, use XPU for training.")
+run_type_g.add_arg("use_cuda",                     bool,   False,  "If set, use GPU for training.")
+run_type_g.add_arg("use_xpu",                      bool,   False,  "If set, use XPU for training.")
 run_type_g.add_arg("use_fast_executor",            bool,   False, "If set, use fast parallel executor (in experiment).")
 run_type_g.add_arg("shuffle",                      bool,   True,  "")
 run_type_g.add_arg("num_iteration_per_drop_scope", int,    1,     "Ihe iteration intervals to clean up temporary variables.")