CoreML model with no inputs fails to load at runtime

[GregoryComer]

🐛 Describe the bug

The following model fails to load at runtime. This is presumably due to not taking inputs? When changing the model to x + torch.randn(...), it loads successfully. This isn't a particularly high-priority case, but does appear to be a bug.

Repro:

import torch

from executorch.backends.apple.coreml.partition import CoreMLPartitioner
from executorch.exir import to_edge_transform_and_lower, EdgeCompileConfig, to_edge
from executorch.extension.pybindings.portable_lib import _load_for_executorch_from_buffer

class Model(torch.nn.Module):
    def __init__(self):
        super().__init__()
        
    def forward(self):
        return torch.randn(5, 5)

model = Model()
inputs = (
)

print(inputs)

eager_outputs = model(*inputs)
print(f"Eager: {eager_outputs.shape} {eager_outputs}")

ep = torch.export.export(model.eval(), inputs)
print(ep)

print(f"EP: {ep.module()(*inputs)}")

lowered = to_edge_transform_and_lower(
    ep,
    partitioner=[CoreMLPartitioner()],
    compile_config=EdgeCompileConfig(_check_ir_validity=False)
).to_executorch()

print(lowered.exported_program())

et_model = _load_for_executorch_from_buffer(lowered.buffer)
et_outputs = et_model([*inputs])[0]

print(et_outputs)

et_outputs - eager_outputs

Outputs:

[ETCoreMLModelManager.mm:566] [Core ML]  Metadata is invalid or missing.
[backend_delegate.mm:288] [Core ML] Model init failed Metadata is invalid or missing.
[coreml_backend_delegate.mm:193] CoreMLBackend: Failed to init the model.
[method.cpp:113] Init failed for backend CoreMLBackend: 0x23

Versions

coremltools version 8.3
executorch commit 67b6009 (Jun 14)

cc @kimishpatel @YifanShenSZ @cymbalrush @metascroy

[metascroy]

Fix for ET here: #13053

Issue re-filed with coremltools here: apple/coremltools#2578