SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1906.00651 · 2019

Probabilistic Noise2Void: Unsupervised Content-Aware Denoising

arXiv · PDF · Open in the Atlas

Code that ran

11 of 19 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 8 have no contract verdict; 3 failures of our instrument; 8 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/1906.00651.json · badge: /api/ran/1906.00651.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
juglab/pn2v community (classifier label)
classifier basis not recorded on the link
10 ran · 3 no recorded run
juglab/ppn2v community (classifier label)
classifier basis not recorded on the link
1 ran · 5 no recorded run
FunctionStatusWhere it lives
PSNR Ran
ran on a synthesized input
juglab/pn2v/src/pn2v/core/utils.py
code served (permissive licence) · get_code("2f7de44d3ac31817")
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)

# PSNR computes peak signal-to-noise ratio between ground truth and prediction
# It uses:
# - gt: array-like, subtractable and squarable
# - pred: array-like, subtractable and squarable
# - range_: float, the dynamic range (default 255.0 for 8-bit images)

# Create small numpy arrays with distinct shapes to test the function
# Ground truth: 2x4 array (batch of 2, 4 values each)
gt = np.random.rand(2, 4) * 255.0

# Prediction: same shape as gt
pred = gt + np.random.randn(2, 4) * 5.0

# Range parameter (typical for 8-bit images)
range_ = 255.0

ARGS = {
    "gt": gt,
    "pred": pred,
    "range_": range_
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, 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
conv1x1 Ran
ran; our contract draft was wrong, not the code
juglab/pn2v/src/pn2v/unet/model.py
code served (permissive licence) · get_code("9200656a2a28834f")
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 Conv2d. 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 torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "in_channels": 3,
    "out_channels": 64,
    "groups": 1,
}
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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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
conv3x3 Ran
ran; our contract draft was wrong, not the code
juglab/pn2v/src/pn2v/unet/model.py
code served (permissive licence) · get_code("fe1bac001ffc64d5")
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 Conv2d. 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 torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'in_channels': 3,
    'out_channels': 16,
    'stride': 1,
    'padding': 1,
    'bias': True,
    'groups': 1,
}
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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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
createHistogram Ran
ran on a synthesized input
juglab/pn2v/src/pn2v/core/histNoiseModel.py
code served (permissive licence) · get_code("8727e74590f791b7")
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 ndarray, ndim 3, dtype family float, shape (3, 4, 4) 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 numpy as np

# Create small 3D numpy arrays for observation and signal
# observation: stack of 2D images (6 images of 3x3)
# signal: stack of 2D images (2 images of 3x3)
# imgFactor will be 6/2 = 3, so 3 observation images per signal image
observation = np.random.rand(6, 3, 3).astype(np.float32)
signal = np.random.rand(2, 3, 3).astype(np.float32)

ARGS = {
    "bins": 4,
    "minVal": 0.0,
    "maxVal": 1.0,
    "observation": observation,
    "signal": signal,
}
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 ndarray, ndim 3, dtype family float, shape (3, 4, 4) — shape (3, 4, 4) 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
getStratifiedCoords2D Ran
ran on a synthesized input
juglab/pn2v/src/pn2v/core/training.py
code served (permissive licence) · get_code("4a437387c73de9d6")
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 list, ndim 2, dtype family int; 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

# Set random seed for reproducibility
np.random.seed(0)

# Define arguments for getStratifiedCoords2D
ARGS = {
    "numPix": 10,
    "shape": (32, 24)
}
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 list, ndim 2, dtype family int — a list 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
imgToTensor Ran
ran on a synthesized input
juglab/pn2v/src/pn2v/core/utils.py
code served (permissive licence) · get_code("49c44b643fa4a445")
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 3, dtype family float, shape (1, 4, 3) 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
import numpy as np
import torchvision

torch.manual_seed(0)

# Create a small 2D single channel image as a numpy array
# The function expects shape (height, width) and will reshape it to (height, width, 1)
img = np.random.rand(4, 3).astype(np.float32)

ARGS = {
    "img": img
}
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 Tensor, ndim 3, dtype family float, shape (1, 4, 3) — shape (1, 4, 3) 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
normalize Ran
ran on a synthesized input
juglab/pn2v/src/pn2v/core/utils.py
code served (permissive licence) · get_code("fe7d72ef6064dfa1")
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 2, dtype family float, shape (2, 4) 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
import numpy as np

torch.manual_seed(0)

# The normalize function performs: (img - mean) / std
# It expects:
# - img: a tensor or array
# - mean: a tensor or array to subtract
# - std: a tensor or array to divide by

img = torch.randn(2, 4)  # batch size 2, 4 dimensions
mean = torch.randn(4)    # per-channel mean
std = torch.randn(4).abs() + 0.1  # per-channel std (ensure non-zero)

ARGS = {
    "img": img,
    "mean": mean,
    "std": std,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
normalize Ran
ran on a synthesized input
juglab/ppn2v/src/ppn2v/pn2v/utils.py
code served (permissive licence) · get_code("583465fbf9bf27ec")
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 4, dtype family float, shape (2, 3, 4, 4) 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
import numpy as np

torch.manual_seed(0)

# Create a small image tensor (batch_size=2, channels=3, height=4, width=4)
img = torch.randn(2, 3, 4, 4)

# Define mean and std as scalars
mean = 0.5
std = 0.2

ARGS = {
    "img": img,
    "mean": mean,
    "std": std
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 3, 4, 4) — shape (2, 3, 4, 4) 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
randomCrop Ran
ran on a synthesized input
juglab/pn2v/src/pn2v/core/training.py
code served (permissive licence) · get_code("5c39a894116876b8")
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

# Set random seed for reproducibility
np.random.seed(0)

# Create small 2D images for testing
# img should be larger than size parameter to allow random cropping
img = np.random.randint(0, 256, size=(64, 64), dtype=np.uint8)

# imgClean should have the same shape as img
imgClean = np.random.randint(0, 256, size=(64, 64), dtype=np.uint8)

# Define ARGS dictionary with all parameters
ARGS = {
    'img': img,
    'size': 32,
    'numPix': 16,
    'imgClean': imgClean,
    'augment': True,
    'manipulate': True
}
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
randomCropFRI Ran
ran on a synthesized input
juglab/pn2v/src/pn2v/core/training.py
code served (permissive licence) · get_code("e88b37fa01225cbf")
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 numpy as np

# Set random seed for reproducibility
np.random.seed(0)

# Create a simple 3D dataset of 2D images
# Stack of 5 images, each 32x32 pixels
data = np.random.randint(0, 256, size=(5, 32, 32), dtype=np.uint8)

# Parameters
size = 16  # patch size (16x16)
numPix = 20  # number of pixels to manipulate
supervised = False
counter = None
augment = True

ARGS = {
    "data": data,
    "size": size,
    "numPix": numPix,
    "supervised": supervised,
    "counter": counter,
    "augment": augment,
}
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 — 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
upconv2x2 Ran
ran; our contract draft was wrong, not the code
juglab/pn2v/src/pn2v/unet/model.py
code served (permissive licence) · get_code("87dc2e4625e5eb8c")
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 ConvTranspose2d. 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 torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'in_channels': 3,
    'out_channels': 16,
    'mode': 'transpose'
}
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 ConvTranspose2d — a ConvTranspose2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ConvTranspose2d 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
getStratifiedCoords2D Not yet run
unverified (harvested, no recorded run)
juglab/ppn2v/src/ppn2v/pn2v/training.py
code served (permissive licence) · get_code("37a52d11132aeea2")
How it ran

no invocation recorded.

predict Not yet run
unverified (harvested, no recorded run)
juglab/pn2v/src/pn2v/core/prediction.py
code served (permissive licence) · get_code("4c5f0a3f4972f81e")
How it ran

no invocation recorded.

randomCrop Not yet run
unverified (harvested, no recorded run)
juglab/ppn2v/src/ppn2v/pn2v/training.py
code served (permissive licence) · get_code("d64d887d53cacd59")
How it ran

no invocation recorded.

randomCropFRI Not yet run
unverified (harvested, no recorded run)
juglab/ppn2v/src/ppn2v/pn2v/training.py
code served (permissive licence) · get_code("16e0687250805927")
How it ran

no invocation recorded.

tiledPredict Not yet run
unverified (harvested, no recorded run)
juglab/pn2v/src/pn2v/core/prediction.py
code served (permissive licence) · get_code("0649cbffc9530d67")
How it ran

no invocation recorded.

tiledPredict Not yet run
unverified (harvested, no recorded run)
juglab/ppn2v/src/ppn2v/pn2v/prediction.py
code served (permissive licence) · get_code("80c9271df6da00fe")
How it ran

no invocation recorded.

tiledPredict_pad Not yet run
unverified (harvested, no recorded run)
juglab/pn2v/src/pn2v/core/prediction.py
code served (permissive licence) · get_code("cd6d147f45d9a060")
How it ran

no invocation recorded.

tiledPredict_pad Not yet run
unverified (harvested, no recorded run)
juglab/ppn2v/src/ppn2v/pn2v/prediction.py
code served (permissive licence) · get_code("580698be9dfa8582")
How it ran

no invocation recorded.

Also present in other papers' code

Of 19 samples lifted from this paper's linked repositories, 19 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 3 later) and 12 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
conv1x1 5 other papers
1 dated earlier, 4 later
1812.10366 (2018-12)
A Poisson-Gaussian Denoising Dataset with Real Fluorescence Microscopy Images
2008.00823 (2020-08)
Rethinking Image Deraining via Rain Streaks and Vapors
2206.02780 (2022-06)
GenSDF: Two-Stage Learning of Generalizable Signed Distance Functions
and 2 more not named here
conv3x3 5 other papers
1 dated earlier, 4 later
1812.10366 (2018-12)
A Poisson-Gaussian Denoising Dataset with Real Fluorescence Microscopy Images
2008.00823 (2020-08)
Rethinking Image Deraining via Rain Streaks and Vapors
2206.02780 (2022-06)
GenSDF: Two-Stage Learning of Generalizable Signed Distance Functions
and 2 more not named here
upconv2x2 4 other papers
1 dated earlier, 3 later
1812.10366 (2018-12)
A Poisson-Gaussian Denoising Dataset with Real Fluorescence Microscopy Images
2206.02780 (2022-06)
GenSDF: Two-Stage Learning of Generalizable Signed Distance Functions
2207.12394 (2022-07)
Dynamic 3D Scene Analysis by Point Cloud Accumulation
and 1 more not named here

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

Today, Convolutional Neural Networks (CNNs) are the leading method for image denoising. They are traditionally trained on pairs of images, which are often hard to obtain for practical applications. This motivates self-supervised training methods such as Noise2Void~(N2V) that operate on single noisy images. Self-supervised methods are, unfortunately, not competitive with models trained on image pairs. Here, we present 'Probabilistic Noise2Void' (PN2V), a method to train CNNs to predict per-pixel intensity distributions. Combining these with a suitable description of the noise, we obtain a complete probabilistic model for the noisy observations and true signal in every pixel. We evaluate PN2V on publicly available microscopy datasets, under a broad range of noise regimes, and achieve competitive results with respect to supervised state-of-the-art methods.

For agents

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

get_harvested_code_for_paper("1906.00651")
get_code_for_paper("1906.00651")
have("1906.00651")

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

curl https://syntology.ai/api/ran/1906.00651.json

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

[![Syntology run record](https://syntology.ai/api/ran/1906.00651.svg)](https://syntology.ai/paper/1906.00651)

Connect an agent — have() is free.