onnx and onnxruntime disagree on input with no known rank

[xadupre]

Describe the bug

The goal is to specify an input with an unknown rank (undefined number of dimensions). Using None as a shape makes check_model fail but onnxruntime works with it.

System information

• OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Windows
• ONNX Runtime installed from (source or binary): source
• ONNX Runtime version: github
• Python version: 3.9
• Visual Studio version (if applicable): 2019

To Reproduce

import numpy
from onnx import numpy_helper, TensorProto, FunctionProto
from onnx.helper import (
    make_model, make_node, set_model_props, make_tensor,
    make_graph, make_tensor_value_info, make_opsetid,
    make_function)
from onnx.checker import check_model
from onnxruntime import InferenceSession


def create_model(shapes):
    new_domain = 'custom'
    opset_imports = [make_opsetid("", 14), make_opsetid(new_domain, 1)]

    node1 = make_node('MatMul', ['X', 'A'], ['XA'])
    node2 = make_node('Add', ['XA', 'A'], ['Y'])

    X = make_tensor_value_info('X', TensorProto.FLOAT, shapes['X'])
    A = make_tensor_value_info('A', TensorProto.FLOAT, shapes['A'])
    Y = make_tensor_value_info('Y', TensorProto.FLOAT, shapes['Y'])

    graph = make_graph([node1, node2], 'example', [X, A], [Y])

    onnx_model = make_model(graph, opset_imports=opset_imports)
    return onnx_model


print("----------- case 1: 2D x 2D -> 2D")
onnx_model = create_model({'X': [None, None], 'A': [None, None], 'Y': [None, None]})
check_model(onnx_model)
sess = InferenceSession(onnx_model.SerializeToString())
res = sess.run(None, {
    'X': numpy.random.randn(2, 2).astype(numpy.float32),
    'A': numpy.random.randn(2, 2).astype(numpy.float32)})
print(res)

print("----------- case 2: 2D x 1D -> 1D")
onnx_model = create_model({'X': [None, None], 'A': [None], 'Y': [None]})
check_model(onnx_model)
sess = InferenceSession(onnx_model.SerializeToString())
res = sess.run(None, {
    'X': numpy.random.randn(2, 2).astype(numpy.float32),
    'A': numpy.random.randn(2).astype(numpy.float32)})
print(res)

print("----------- case 3: 2D x 0D -> 0D")
onnx_model = create_model({'X': [None, None], 'A': [], 'Y': []})
check_model(onnx_model)
try:
    InferenceSession(onnx_model.SerializeToString())
except Exception as e:
    print(e)

print("----------- case 4: 2D x None -> None")
onnx_model = create_model({'X': [None, None], 'A': None, 'Y': None})
try:
    check_model(onnx_model)
except Exception as e:
    print(type(e), e)
sess = InferenceSession(onnx_model.SerializeToString())
res = sess.run(None, {
    'X': numpy.random.randn(2, 2).astype(numpy.float32),
    'A': numpy.random.randn(2).astype(numpy.float32)})
print(res)
print("----------- end")

Output:

----------- case 1: 2D x 2D -> 2D
[array([[-2.7603245 , -3.3140254 ],
       [ 0.8899558 ,  0.50718915]], dtype=float32)]
----------- case 2: 2D x 1D -> 1D
[array([-1.242917  ,  0.41560578], dtype=float32)]
----------- case 3: 2D x 0D -> 0D
[ONNXRuntimeError] : 1 : FAIL : Node () Op (MatMul) [ShapeInferenceError] Input tensors of wrong rank (0).
----------- case 4: 2D x None -> None
<class 'onnx.onnx_cpp2py_export.checker.ValidationError'> Field 'shape' of type is required but missing.
[array([-1.7270846,  1.0475622], dtype=float32)]
----------- end

Expected behavior
A way to define an input with an undefined number of dimensions.

[microsoft-github-policy-service]

Applying stale label due to no activity in 30 days