SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2110.12271 · 2021

Self-Validation: Early Stopping for Single-Instance Deep Generative Priors

arXiv · PDF · Open in the Atlas

Code that ran

5 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 0 failures of our instrument; 4 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2110.12271.json · badge: /api/ran/2110.12271.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
sun-umn/Self-Validation community (classifier label)
classifier basis not recorded on the link
5 ran · 4 no recorded run
FunctionStatusWhere it lives
Pseudo_Huber_Loss Ran
ran on a synthesized input
sun-umn/Self-Validation/DD+AE/Net512/W256_P500/util.py
code served (permissive licence) · get_code("ee805b2cebc4c91c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create small tensors for true_data and pred_data
# Both should be the same shape for element-wise operations
true_data = torch.randn(2, 4)
pred_data = torch.randn(2, 4)

# delta is a scalar threshold parameter
delta = 1.0

# device is CPU
device = torch.device('cpu')

ARGS = {
    'true_data': true_data,
    'pred_data': pred_data,
    'delta': delta,
    'device': device
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
calcualte_PSNR Ran
ran on a synthesized input
sun-umn/Self-Validation/DD+AE/Net512/W256_P500/psnr_ssim.py
code served (permissive licence) · get_code("6911cf7209ee6f23")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import torch

torch.manual_seed(0)

# The function expects two image arrays with shape (C, H, W)
# that will be transposed to (H, W, C) for PSNR calculation
# We need small arrays with distinct dimensions for clarity

# Create two small image arrays with shape (C, H, W)
# C=3 (channels), H=4 (height), W=5 (width)
im_true = np.random.randint(0, 256, size=(3, 4, 5), dtype=np.uint8)
im_test = np.random.randint(0, 256, size=(3, 4, 5), dtype=np.uint8)

ARGS = {
    "im_true": im_true,
    "im_test": im_test,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, skimage, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
create_clean_corrupted_dataset Ran
ran on a synthesized input
sun-umn/Self-Validation/DD+AE/Net512/W256_P500/util.py
code served (permissive licence) · get_code("8c7a7b681dec6a9b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import tempfile
import os

# Create temporary directory for test files
temp_dir = tempfile.mkdtemp()

# Create clean data file
clean_data = np.array([[1.0, 2.0, 3.0, 4.0],
                       [5.0, 6.0, 7.0, 8.0]])
clean_file = os.path.join(temp_dir, "clean.npz")
np.savez(clean_file, clean_data)

# Create corrupted data file
corrupted_data = np.array([[1.1, 2.1, 3.1, 4.1],
                           [5.1, 6.1, 7.1, 8.1]])
corrupted_file = os.path.join(temp_dir, "corrupted.npz")
np.savez(corrupted_file, corrupted_data)

ARGS = {
    "clean_file": clean_file,
    "corrupted_file": corrupted_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 3, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
prepare_AEData Ran
ran on a synthesized input
sun-umn/Self-Validation/DD+AE/Net512/W256_P500/AE_util.py
code served (permissive licence) · get_code("d210bccf2036f01e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
import torchvision
from torch.utils.data import Dataset, DataLoader

torch.manual_seed(0)

# Create small dummy input data
# The function expects input_data to be array-like with shape (num, features)
# AEDataset indexes it with [idx, :], so it needs to support 2D indexing
input_data = np.random.randn(6, 4).astype(np.float32)

batch_size = 2
num = 6

ARGS = {
    "input_data": input_data,
    "batch_size": batch_size,
    "num": num,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
prepare_data Ran
ran on a synthesized input
sun-umn/Self-Validation/DD+AE/Net512/W256_P500/util.py
code served (permissive licence) · get_code("6fb6eb7b9a9f460f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned DataLoader. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
import tempfile
import os

torch.manual_seed(0)

# Create temporary npz files with small dummy data
temp_dir = tempfile.mkdtemp()

# Create small dummy arrays (3x4x5 for height x width x channels or similar)
clean_array = np.random.randint(0, 256, (3, 4, 5), dtype=np.uint8)
corrupted_array = np.random.randint(0, 256, (3, 4, 5), dtype=np.uint8)

clean_file = os.path.join(temp_dir, "clean.npz")
corrupted_file = os.path.join(temp_dir, "corrupted.npz")

np.savez(clean_file, clean_array)
np.savez(corrupted_file, corrupted_array)

ARGS = {
    "clean_image_path": clean_file,
    "corrupted_image_path": corrupted_file,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataLoader — a DataLoader output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataLoader output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
DD_test Not yet run
unverified (harvested, no recorded run)
sun-umn/Self-Validation/DD+AE/Net512/W256_P500/train_denoising.py
code served (permissive licence) · get_code("7009ba0103223c36")
How it ran

no invocation recorded.

SmoothIt Not yet run
unverified (harvested, no recorded run)
sun-umn/Self-Validation/DD+AE/Net512/W256_P500/Smooth_Values.py
code served (permissive licence) · get_code("64e1e4bcbdc82ceb")
How it ran

no invocation recorded.

autoencoder_test Not yet run
unverified (harvested, no recorded run)
sun-umn/Self-Validation/DD+AE/Net512/W256_P500/AE_bp.py
code served (permissive licence) · get_code("e0e4aa9f9075cb4a")
How it ran

no invocation recorded.

train_stopAE Not yet run
unverified (harvested, no recorded run)
sun-umn/Self-Validation/DD+AE/Net512/W256_P500/AE_train.py
code served (permissive licence) · get_code("6fca8ea7b6cf9d27")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 8 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
calcualte_PSNR 1 other paper
1 later
Li_Deep_Random_Projector_Accelerated_Deep_Image_Prior_CVPR_2023_paper (2023 (year only))
Deep Random Projector: Accelerated Deep Image Prior

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Recent works have shown the surprising effectiveness of deep generative models in solving numerous image reconstruction (IR) tasks, even without training data. We call these models, such as deep image prior and deep decoder, collectively as single-instance deep generative priors (SIDGPs). The successes, however, often hinge on appropriate early stopping (ES), which by far has largely been handled in an ad-hoc manner. In this paper, we propose the first principled method for ES when applying SIDGPs to IR, taking advantage of the typical bell trend of the reconstruction quality. In particular, our method is based on collaborative training and self-validation: the primal reconstruction process is monitored by a deep autoencoder, which is trained online with the historic reconstructed images and used to validate the reconstruction quality constantly. Experimentally, on several IR problems and different SIDGPs, our self-validation method is able to reliably detect near-peak performance and signal good ES points. Our code is available at https://sun-umn.github.io/Self-Validation/.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2110.12271")
get_code_for_paper("2110.12271")
have("2110.12271")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2110.12271.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2110.12271.svg)](https://syntology.ai/paper/2110.12271)

Connect an agent — have() is free.