doc(hansbug): create a new start up

docs/source/index.rst

@@ -13,6 +13,7 @@ module.
     :caption: Tutorials
 
     tutorials/installation/index
+    tutorials/quick_start/index
 
 
 .. toctree::

docs/source/tutorials/quick_start/create_tensor.demo.py

+import builtins
+import os
+from functools import partial
+
+import treetensor.torch as torch
+
+print = partial(builtins.print, sep=os.linesep)
+
+if __name__ == '__main__':
+    t1 = torch.tensor([[1, 2, 3],
+                       [4, 5, 6]])
+    print('new native tensor:', t1)
+
+    t2 = torch.tensor({
+        'a': [1, 2, 3],
+        'b': {'x': [[4, 5], [6, 7]]},
+    })
+    print('new tree tensor:', t2)
+
+    t3 = torch.randn(2, 3)
+    print('new random native tensor:', t3)
+
+    t4 = torch.randn({
+        'a': (2, 3),
+        'b': {'x': (3, 4)},
+    })
+    print('new random tree tensor:', t4)

docs/source/tutorials/quick_start/index.rst

+Quick Start
+==================
+
+Create a Tree-based Tensor
+------------------------------
+
+You can create a tree-based tensor or a native tensor \
+like the following example code.
+
+.. literalinclude:: create_tensor.demo.py
+    :language: python
+    :linenos:
+
+The output should be like below.
+
+.. literalinclude:: create_tensor.demo.py.txt
+    :language: text
+    :linenos:
+