Refactor IO config to yaml in assets

[xiaoyu-work]

This pull request introduces a major refactor of the HuggingFace ONNX I/O configuration system in Olive. The changes move away from hardcoded Python class registries and instead adopt a YAML-driven approach for specifying input/output configurations for both tasks and diffusers pipelines. This enables easier extensibility and clearer, data-driven configuration management. Additionally, the update adds default dummy input shapes and rich configuration templates for tasks and diffusers components.

The most important changes are:

YAML-based IO Configuration System:

• Added new YAML files (defaults.yaml, tasks.yaml, diffusers.yaml) in olive/assets/io_configs/ that define default dummy input shapes, task templates, and diffusers component/pipeline specifications for ONNX export. These files replace the previous Python-based registries and provide a more flexible and maintainable configuration system. [1] [2] [3]

Refactor of IO Config Python Interface:

• Replaced the contents of olive/common/hf/io_config/__init__.py to remove all hardcoded class registries and mapping logic, and instead import new YAML-driven utility functions (generate_dummy_inputs, get_io_config, get_diffusers_io_config, etc.). This greatly simplifies the interface and shifts responsibility to the new YAML configuration system.

Legal and Organizational Improvements:

• Added copyright and license headers to olive/assets/__init__.py and olive/assets/io_configs/__init__.py.## Describe your changes

Checklist before requesting a review

• Add unit tests for this change.
• Make sure all tests can pass.
• Update documents if necessary.
• Lint and apply fixes to your code by running lintrunner -a
• Is this a user-facing change? If yes, give a description of this change to be included in the release notes.

(Optional) Issue link

[Copilot]

Pull request overview

This pull request introduces a major refactor of the HuggingFace ONNX I/O configuration system, migrating from hardcoded Python class registries to a flexible YAML-driven approach. The changes enable easier extensibility and clearer configuration management for model optimization workflows.

Changes:

• Replaced Python-based OnnxConfig class registries with YAML configuration files for tasks and diffusers components
• Introduced unified dummy input generators that read specifications from YAML rather than class hierarchies
• Consolidated I/O configuration generation into a single task-driven API with automatic optional input filtering based on model signatures

Reviewed changes

Copilot reviewed 29 out of 34 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
olive/assets/io_configs/tasks.yaml	Defines input/output specifications for 15+ HuggingFace tasks
olive/assets/io_configs/diffusers.yaml	Defines I/O specs for diffusers components (SD, SDXL, SD3, Flux, Sana)
olive/assets/io_configs/defaults.yaml	Provides default dimension values and attribute aliases for config resolution
olive/common/hf/io_config/yaml_loader.py	YAML loading and config resolution utilities with LRU caching
olive/common/hf/io_config/task_config.py	New unified API for generating I/O configs and dummy inputs
olive/common/hf/io_config/input_generators.py	Refactored to use YAML-driven input generation
olive/common/hf/model_io.py	Updated to use new task-driven API
olive/common/hf/wrapper.py	Migrated to use resolve_alias for config attribute access
test/common/hf/io_config/*.py	Updated tests for new YAML-driven system

Comments suppressed due to low confidence (1)

olive/common/hf/io_config/task_config.py:1

• Missing check for whether 'with_past' key exists in task_spec before accessing it. This will raise a KeyError if a task doesn't define with_past overrides. Add a check: if self.use_past_in_inputs and 'with_past' in self.task_spec:

# -------------------------------------------------------------------------