First sketch of a very simple API to create simple graphs (#42)

* First sketch of a very simple API to create simple graphs

* doc

* f401

* add python-re2

* add re2

* update

* doc

* array

* almost done

* doc

* add many ops

* doc

* doc

Author: xadupre

CHANGELOGS.rst

@@ -4,6 +4,7 @@ Change Logs
 0.2.0
 +++++
 
+* :pr:`42`: first sketch for a very simple API to create onnx graph in one or two lines
 * :pr:`27`: add function from_array_extended to convert
   an array to a TensorProto, including bfloat16 and float 8 types
 * :pr:`24`: add ExtendedReferenceEvaluator to support scenario

README.rst

@@ -114,5 +114,4 @@ It supports eager mode as well:
 The library is released on
 `pypi/onnx-array-api <https://pypi.org/project/onnx-array-api/>`_
 and its documentation is published at
-`(Numpy) Array API for ONNX
-<https://sdpython.github.io/doc/onnx-array-api/dev/>`_.
+`(Numpy) Array API for ONNX <https://sdpython.github.io/doc/onnx-array-api/dev/>`_.

_doc/api/index.rst

@@ -7,6 +7,7 @@ API
     :maxdepth: 1
 
     array_api
+    light_api
     npx_core_api
     npx_functions
     npx_jit_eager

_doc/api/light_api.rst

@@ -0,0 +1,32 @@
+========================
+onnx_array_api.light_api
+========================
+
+start
+=====
+
+.. autofunction:: onnx_array_api.light_api.start
+
+OnnxGraph
+=========
+
+.. autoclass:: onnx_array_api.light_api.OnnxGraph
+    :members:
+
+BaseVar
+=======
+
+.. autoclass:: onnx_array_api.light_api.var.BaseVar
+    :members:
+
+Var
+===
+
+.. autoclass:: onnx_array_api.light_api.Var
+    :members:
+
+Vars
+====
+
+.. autoclass:: onnx_array_api.light_api.Vars
+    :members: