Public

Ptychography — Public Tier

(5 scenes)

Full-access development tier with all data visible.

What you get

Measurements (y), ideal forward operator (H), spec ranges, ground truth (x_true), and true mismatch spec.

How to use

Load HDF5 → compare reconstruction vs x_true → check consistency → iterate.

What to submit

Reconstructed signals (x_hat) and corrected spec as HDF5.

Parameter Specifications

True spec visible — use these exact values for Scenario III oracle reconstruction.

Parameter Spec Range True Value Unit
probe_error -5.0 – 10.0 2.5 %
position_error -10.0 – 20.0 5.0 nm
partial_coherence -5.0 – 10.0 2.5 nm

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

7.15 dB

SSIM 0.4195

Scenario II (Mismatch)

8.74 dB

SSIM 0.6343

Scenario III (Oracle)

2.11 dB

SSIM 0.0191

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 10.66 0.5537 8.05 0.5494 3.68 0.0192
scene_01 8.28 0.5839 7.32 0.4150 1.97 0.0202
scene_02 4.88 0.0605 9.53 0.8318 1.14 0.0113
scene_03 4.78 0.4799 10.06 0.7411 1.65 0.0258

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 AutoPhaseNN + gradient 0.770 31.95 0.942 0.87 ✓ Certified Chan et al., Commun. Phys. 2024
2 PtychoNN + gradient 0.745 29.87 0.915 0.9 ✓ Certified Cherukara et al., Appl. Phys. Lett. 2020
3 sDR + gradient 0.700 26.92 0.856 0.94 ✓ Certified Wen et al., J. Opt. 2019
4 ePIE + gradient 0.590 22.61 0.715 0.89 ✓ Certified Maiden & Rodenburg, Ultramicroscopy 2009

Visible Data Fields

y H_ideal spec_ranges x_true true_spec

Dataset

Format: HDF5
Scenes: 5

Scoring Formula

0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

PSNR: 40% SSIM: 40% Consistency: 20%
Back to Ptychography