Fix Travis Ci build failed

* Also Update some examples of to_excel method

Author: newinh

pandas/core/generic.py

@@ -1917,31 +1917,23 @@ def _repr_latex_(self):
     ...                   columns=['col 1', 'col 2'])
     >>> df1.to_excel("output.xlsx")
 
+    If you want to set engine that can manipulate Excel,
+    pass keyword argument named engine. Actually
+    engine is automatically chosen by file extension:
+
+    >>> df1.to_excel('output1.xlsx', engine='xlsxwriter')
+
     To specify the sheet name:
 
     >>> df1.to_excel("output.xlsx", sheet_name='Sheet_name_1')
 
     If you wish to write to more than one sheet in the workbook, it is
     necessary to specify an ExcelWriter object:
 
-    >>> with ExcelWriter('output.xlsx') as writer:
+    >>> with pd.ExcelWriter('output.xlsx') as writer:
     ...     df1.to_excel(writer, sheet_name='Sheet_name_1')
     ...     df2.to_excel(writer, sheet_name='Sheet_name_2')
 
-    If you want to set engine that can manipulate Excel,
-    pass keyword argument named engine. Actually
-    engine is automatically chosen by file extension:
-
-    >>> df1.to_excel('output1.xlsx', engine='xlsxwriter')
-    >>> with ExcelWriter('output2.xlsx', engine='openpyxl') as writer:
-    ...     df2.to_excel(writer)
-
-    You can set date format or datetime format:
-
-    >>> with ExcelWriter('path_to_file.xlsx',
-                          date_format='YYYY-MM-DD',
-                          datetime_format='YYYY-MM-DD HH:MM:SS') as writer:
-    ...     df.to_excel(writer)
 
     """