[Bug][Relax][ONNX] Split._impl_v13 rejects graph-initializer split operand under keep_params_in_input=True

[zacharywhitley]

Summary

Split._impl_v13 errors on a splits operand that arrived as a
relax.Var bound to a graph initializer under
keep_params_in_input=True. The data is fully static; the isinstance
check is too strict.

Environment

TVM 0.25.0.post1 (pip), macOS arm64, Python 3.11.

Reproducer

Model: owensong/Inflect-Nano-v2 on Hugging Face (Apache-2.0).

import onnx, onnxsim
from tvm.relax.frontend.onnx import from_onnx

model = onnx.load("encoder.onnx")
model, _ = onnxsim.simplify(
    model,
    overwrite_input_shapes={"tokens": [1, 64], "lengths": [1], "length_scale": []},
)
mod = from_onnx(model, keep_params_in_input=True)  # the recommended path

Failure

ValueError: Dynamic Split not yet supported

At python/tvm/relax/frontend/onnx/onnx_frontend.py:2378.
Under keep_params_in_input=True, initializers become relax.Vars
rather than relax.Constants. The check
isinstance(splits, relax.Constant) fails even when the Var is
bound to a relax.const.

Suggested patch

Follow a Var back to its constant binding before the isinstance
check.

--- a/python/tvm/relax/frontend/onnx/onnx_frontend.py
+++ b/python/tvm/relax/frontend/onnx/onnx_frontend.py
@@ Split._impl_v13
+        # Under keep_params_in_input=True, splits may be a Var bound to a Constant.
+        if isinstance(splits, relax.Var):
+            binding = bb.lookup_binding(splits)
+            if isinstance(binding, relax.Constant):
+                splits = binding
         if not isinstance(splits, relax.Constant):
             raise ValueError("Dynamic Split not yet supported")

(bb.lookup_binding is illustrative — the exact spelling depends on
which builder API is in scope at that point in the frontend.)

User workaround

keep_params_in_input=False — initializers become relax.Constants
directly, satisfying the isinstance check. Trade-off: params fold
into the module body rather than being module-level Vars, changing
downstream analyses.

Discovered by

Compiling Inflect nano encoder from
cognition. See sibling
reports for two related bugs (bug 3 has no user workaround).