functional-conversion

Name: functional-conversion
Availability: InStock
Author: just-it

Converts PyTorch MLP models to graph-format ONNX with pluggable backends and optimizers.

33|51|Updated Mar 12, 2026

Search Tags:#pytorch #convert #functional #code-transformation #dsl-generation #nn.module #model-rewrite

Ready to run

npx skills add https://github.com/Just-it/AscendOpGenAgent --skill functional-conversion

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: functional-conversion
Source: https://github.com/Just-it/AscendOpGenAgent/tree/main/skills/functional_conversion
Command: npx skills add https://github.com/Just-it/AscendOpGenAgent --skill functional-conversion

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Converts PyTorch nn.Module classes to functional API style, removing class wrappers and self parameters to produce a concise, testable functional implementation for DSL generation.

Core Features & Use Cases

Functionalizes modules by replacing Module.forward calls with a module_fn that uses torch.nn.functional operations, preserving behavior.
Generates a minimal, testable functional model including imports, a Model(nn.Module) wrapper, and helper input getters for quick validation.
Enables DSL-based code generation, deployment, and reproducibility by providing a consistent functional interface across different ops via the provided references.

Quick Start

Convert a PyTorch nn.Module reference and its related references into a functional PyTorch implementation suitable for DSL pipelines.

Frequently Asked Questions about functional-conversion

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I convert PyTorch nn.Module to functional code for DSL generation?▼

You convert PyTorch nn.Module to functional code by removing class wrappers and replacing forward calls with torch.nn.functional operations via a module_fn, generating a minimal functional implementation suitable for DSL generation.

What is the functional API equivalent in PyTorch and when do I need it?▼

A PyTorch functional API equivalent replaces class-based nn.Module definitions with functional calls using torch.nn.functional operations, needed when simplifying model deployment and ensuring reproducibility in DSL environments.

Does converting nn.Module to a functional model preserve the original behavior?▼

Yes, converting nn.Module to a functional model preserves behavior and structure by replacing Module.forward calls with equivalent torch.nn.functional operations while removing the class-based wrapper.

Can I rewrite a PyTorch model to remove class wrappers for deployment?▼

Yes, you can rewrite a PyTorch model to remove class wrappers by generating a functional forward path using module_fn and functional calls, which directly supports model deployment in DSL environments.

What are the limitations of functionalizing PyTorch modules for DSL environments?▼

Functionalizing PyTorch modules requires reading input references and category-specific examples to generate the functional forward path, meaning modules lacking clear reference examples may face conversion limitations.