feat(keypoint-detection): enable ViTPose config/build/perf

src/winml/modelkit/export/htp/exporter.py

@@ -229,7 +229,13 @@ def export(
             monitor.update(ExportStep.INPUT_GEN, **input_gen_data)
 
             # Step 3: Hierarchy Building
-            self._trace_model_hierarchy(model, inputs)
+            # Trace under the Optimum patcher so models that inject constant
+            # forward arguments at export time (e.g. ViTPose MoE's dataset_index)
+            # are traced with the same inputs they are exported with. The export
+            # in Step 4 re-enters the patcher; the contexts are sequential, not
+            # nested.
+            with self._get_optimum_patcher(model, task):
+                self._trace_model_hierarchy(model, inputs)
 
             execution_steps = (
                 self._hierarchy_builder.get_execution_summary().get("execution_steps", 0)
@@ -487,7 +493,11 @@ def _get_optimum_patcher(model: nn.Module, task: str | None) -> Any:
                 task=to_optimum_task(task),
                 library_name="transformers",
             )
-            return cfg_cls(model_config).patch_model_for_export(model)
+            # Pass an explicit empty model_kwargs so patchers that inject extra
+            # forward arguments can populate it. Some patchers (e.g. ViTPose MoE,
+            # which sets a constant dataset_index) assume a mutable dict and crash
+            # on the None default from patch_model_for_export.
+            return cfg_cls(model_config).patch_model_for_export(model, model_kwargs={})
         except KeyError:
             logger.debug(
                 "Model type '%s' (task='%s') not in Optimum registry; "

src/winml/modelkit/models/hf/__init__.py

@@ -75,6 +75,7 @@
     VisionDecoderIOConfig as _VisionDecoderIOConfig,  # triggers registration
 )
 from .vision_encoder_decoder import VisionEncoderIOConfig as _VisionEncoderIOConfig
+from .vitpose import MODEL_CLASS_MAPPING as _VITPOSE_CLASS_MAPPING
 from .zoedepth import ZoeDepthIOConfig as _ZoeDepthIOConfig  # triggers registration
 
 
@@ -97,6 +98,7 @@
     **_SIGLIP_CLASS_MAPPING,
     **_T5_CLASS_MAPPING,
     **_VED_CLASS_MAPPING,
+    **_VITPOSE_CLASS_MAPPING,
 }
 
 # Registry: model_type -> WinMLBuildConfig

src/winml/modelkit/models/hf/vitpose.py

@@ -0,0 +1,48 @@
+# -------------------------------------------------------------------------
+# Copyright (c) Microsoft Corporation. All rights reserved.
+# Licensed under the MIT License.
+# --------------------------------------------------------------------------
+"""ViTPose HuggingFace Model Configuration.
+
+ViTPose is a top-down human pose (keypoint-detection) model: a plain ViT
+backbone with a lightweight decoder that regresses keypoint heatmaps inside a
+given person box.
+
+This module provides:
+- MODEL_CLASS_MAPPING: routes keypoint-detection to VitPoseForPoseEstimation,
+  and declares it the default task via a (vitpose, None) sentinel.
+
+Why ViTPose needs class mapping:
+Optimum already registers the ONNX export config (VitPoseOnnxConfig) for the
+"vitpose" model type, so export works once the model is loaded. However,
+Optimum's TasksManager has no task-to-class entry for "keypoint-detection",
+and transformers' AutoModelForKeypointDetection only recognizes SuperPoint —
+not ViTPose. Without this mapping the resolver cannot load the model class for
+the keypoint-detection task. The "plus" checkpoints (MoE backbone) load through
+the same class; their expert index is fixed at export time by Optimum's
+VitPoseModelPatcher, so no extra input is needed.
+
+Why the (vitpose, None) sentinel:
+TasksManager cannot infer a task from the ViTPose architecture, so without a
+declared default the resolver falls back to an unrelated task and config/build
+fail unless the user passes --task keypoint-detection. The sentinel encodes
+keypoint-detection as the canonical default (the resolver reverse-looks-up the
+task sharing the sentinel's class), making --task optional. Mirrors SAM, which
+declares mask-generation the same way.
+"""
+
+from __future__ import annotations
+
+from transformers import VitPoseForPoseEstimation
+
+
+# (model_type, task) -> HuggingFace model class
+#
+# The (vitpose, None) sentinel declares keypoint-detection as the default task
+# applied during auto-detection (when the user does not pass --task). Its value
+# is the default *class*; the resolver reverse-looks-up the task name from the
+# matching (vitpose, keypoint-detection) -> same class entry.
+MODEL_CLASS_MAPPING: dict[tuple[str, str | None], type] = {
+    ("vitpose", "keypoint-detection"): VitPoseForPoseEstimation,
+    ("vitpose", None): VitPoseForPoseEstimation,
+}

tests/unit/export/test_htp_exporter_patcher_model_kwargs.py

@@ -0,0 +1,73 @@
+# -------------------------------------------------------------------------
+# Copyright (c) Microsoft Corporation. All rights reserved.
+# Licensed under the MIT License.
+# --------------------------------------------------------------------------
+"""Regression tests for `HTPExporter._get_optimum_patcher` model_kwargs handling.
+
+Some Optimum model patchers populate a mutable ``model_kwargs`` dict to inject
+constant forward arguments at export time. ViTPose's MoE patcher, for example,
+sets ``model_kwargs["dataset_index"]`` when ``num_experts > 1``. Optimum's
+``patch_model_for_export`` defaults ``model_kwargs`` to ``None``, so such
+patchers crash with ``TypeError: 'NoneType' object does not support item
+assignment`` unless the caller passes an explicit dict.
+
+This test pins the contract that ``_get_optimum_patcher`` passes an explicit
+``model_kwargs={}`` so those patchers can populate it.
+"""
+
+from __future__ import annotations
+
+from unittest.mock import MagicMock, patch
+
+import torch.nn as nn
+
+from winml.modelkit.export.htp import HTPExporter
+
+
+class _FakeConfig:
+    """Minimal HF-style config exposing the model_type the patcher checks."""
+
+    model_type = "vitpose"
+
+
+class _FakeModel(nn.Module):
+    def __init__(self) -> None:
+        super().__init__()
+        self.config = _FakeConfig()
+
+
+class TestGetOptimumPatcherModelKwargs:
+    """_get_optimum_patcher must pass an explicit mutable model_kwargs dict."""
+
+    def test_patch_model_for_export_receives_explicit_dict(self) -> None:
+        """The patcher call must pass ``model_kwargs={}`` (not the None default).
+
+        We patch the TasksManager lookup to return a fake config constructor
+        whose ``patch_model_for_export`` records the ``model_kwargs`` it
+        receives. A non-None dict lets MoE patchers populate forward arguments
+        without crashing.
+        """
+        captured: dict[str, object] = {}
+
+        fake_onnx_config = MagicMock()
+
+        def record_patch(model, model_kwargs=None):
+            captured["model_kwargs"] = model_kwargs
+            return MagicMock()
+
+        fake_onnx_config.patch_model_for_export.side_effect = record_patch
+
+        def fake_ctor(*args: object, **kwargs: object):
+            return fake_onnx_config
+
+        with patch(
+            "optimum.exporters.tasks.TasksManager.get_exporter_config_constructor",
+            return_value=fake_ctor,
+        ):
+            HTPExporter._get_optimum_patcher(_FakeModel(), task="keypoint-detection")
+
+        assert captured.get("model_kwargs") == {}, (
+            "Expected _get_optimum_patcher to pass an explicit model_kwargs={} "
+            f"to patch_model_for_export, got {captured.get('model_kwargs')!r}. "
+            "MoE patchers (e.g. ViTPose dataset_index) need a mutable dict."
+        )

tests/unit/models/test_vitpose_mapping.py

@@ -0,0 +1,69 @@
+# -------------------------------------------------------------------------
+# Copyright (c) Microsoft Corporation. All rights reserved.
+# Licensed under the MIT License.
+# --------------------------------------------------------------------------
+"""Tests for ViTPose keypoint-detection model-class resolution.
+
+Optimum registers the ViTPose ONNX export config but has no
+task-to-class entry for ``keypoint-detection``, and transformers'
+``AutoModelForKeypointDetection`` only recognises SuperPoint. The
+``("vitpose", "keypoint-detection")`` entry in ``MODEL_CLASS_MAPPING``
+bridges that gap so the resolver can load ``VitPoseForPoseEstimation``.
+"""
+
+from __future__ import annotations
+
+from unittest.mock import MagicMock
+
+from winml.modelkit.loader import resolve_task
+from winml.modelkit.models.hf import MODEL_CLASS_MAPPING
+from winml.modelkit.models.hf.vitpose import MODEL_CLASS_MAPPING as VITPOSE_MAPPING
+
+
+class TestVitPoseMapping:
+    """ViTPose keypoint-detection routes to VitPoseForPoseEstimation."""
+
+    def test_mapping_entry_registered(self):
+        """The aggregated mapping exposes the vitpose keypoint-detection entry."""
+        assert ("vitpose", "keypoint-detection") in MODEL_CLASS_MAPPING
+        assert (
+            MODEL_CLASS_MAPPING[("vitpose", "keypoint-detection")].__name__
+            == "VitPoseForPoseEstimation"
+        )
+
+    def test_module_mapping_merged_into_aggregate(self):
+        """The module-level mapping is included in the aggregated mapping."""
+        assert VITPOSE_MAPPING.items() <= MODEL_CLASS_MAPPING.items()
+
+    def test_explicit_task_resolves_vitpose_class(self):
+        """An explicit keypoint-detection task resolves VitPoseForPoseEstimation."""
+        config = MagicMock()
+        config.model_type = "vitpose"
+        config.architectures = ["VitPoseForPoseEstimation"]
+        config._name_or_path = "usyd-community/vitpose-base-simple"
+
+        resolution = resolve_task(config, task="keypoint-detection")
+
+        assert resolution.task == "keypoint-detection"
+        assert resolution.model_class.__name__ == "VitPoseForPoseEstimation"
+
+    def test_sentinel_resolves_default_task_without_explicit_task(self):
+        """With no --task, the (vitpose, None) sentinel defaults to keypoint-detection."""
+        config = MagicMock()
+        config.model_type = "vitpose"
+        config.architectures = ["VitPoseForPoseEstimation"]
+        config._name_or_path = "usyd-community/vitpose-base-simple"
+
+        resolution = resolve_task(config)
+
+        assert resolution.task == "keypoint-detection"
+        assert resolution.model_class.__name__ == "VitPoseForPoseEstimation"
+
+    def test_sentinel_registered_in_mapping(self):
+        """The (vitpose, None) sentinel shares the keypoint-detection class."""
+        assert ("vitpose", None) in MODEL_CLASS_MAPPING
+        assert (
+            MODEL_CLASS_MAPPING[("vitpose", None)]
+            is MODEL_CLASS_MAPPING[("vitpose", "keypoint-detection")]
+        )
+