Skip to content

v6.2.1 ablation: RepVGG/ECBSR/RepSR reparameterization for student backbone #9

Description

@cashcon57

From external feedback (2026-05-08): consider RepVGG-style (or ECBSR/RepSR) multi-branch reparameterizable training in the student backbone — train with 3×3 + 1×1 + identity branches in parallel, fuse to a single 3×3 at inference for zero added inference cost.

Current state: student backbone (oss/sr/v6/student/, commit aba1fc8) is FasterNet-based per FSR-4 architectural observations memo. FasterNet was chosen because AMD's MIT-licensed FSR 4 source validates it on shipping hardware.

This issue: AFTER v6.2-pico-002 baseline converges, ablate three configurations:

  1. Current FasterNet baseline
  2. FasterNet with RepBlock inside the 1×1 expand path (non-standard hybrid)
  3. Pure RepVGG-style stack (drops FasterNet partial-conv pattern)

Quality target: +0.1-0.3 dB PSNR vs FasterNet baseline at same inference cost. If hybrid (#2) wins, consider for v6.2-pico-002.2.

Refs:

  • RepVGG: Ding et al., CVPR 2021
  • ECBSR: edge-oriented conv block for SR (mobile)
  • RepSR: ESRGAN-family reparameterization

Companion: docs/research/2026-05-08-fsr4-architecture-observations.md

Metadata

Metadata

Assignees

No one assigned

    Labels

    distillationModel distillationpriority-p1Tier 2 priorityv6.2v6.2 architecture work

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions