SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1502.04623 · 2015

DRAW: A Recurrent Neural Network For Image Generation

arXiv · PDF · Open in the Atlas

Code that ran

9 of 18 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, 1 violated it and 2 have no contract verdict; 6 failures of our instrument; 9 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/1502.04623.json · badge: /api/ran/1502.04623.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
suhoy901/DRAW_pytorch reimplementation (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
csiki/v2a extension (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
MoustafaMeshry/draw community (classifier label)
classifier basis not recorded on the link
1 ran · 3 no recorded run
jbornschein/draw community (classifier label)
classifier basis not recorded on the link
1 ran · 2 no recorded run
copy not recorded community
not in the paper's declared repositories
1 ran · 0 no recorded run
ericjang/draw reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
jlindsey15/DRAM reimplementation (classifier label)
classifier basis: classifier_2025
0 ran · 3 no recorded run
singhsidhukuldeep/Generating-Devanagari-Using-DRAW community (classifier label)
classifier basis not recorded on the link
0 ran · 1 no recorded run
FunctionStatusWhere it lives
Variable Ran
ran; our contract draft was wrong, not the code
suhoy901/DRAW_pytorch/utils.py
pointer only (licence: NONE) · get_code("fa0f1cbd96734e5f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, suhoy901/DRAW_pytorch when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

align Ran
ran; our contract draft was wrong, not the code
suhoy901/DRAW_pytorch/utils.py
pointer only (licence: NONE) · get_code("13490b546e6d33df")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, suhoy901/DRAW_pytorch when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

find_model Ran
ran; our contract draft was wrong, not the code
csiki/v2a/config.py
pointer only (licence: NONE) · get_code("330f6334fb6a3c24")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, csiki/v2a when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

load_config Ran
ran; our contract draft was wrong, not the code
csiki/v2a/config.py
pointer only (licence: NONE) · get_code("5e0281ad454dfda9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, csiki/v2a when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

scale_norm Ran
ran on a synthesized input
jbornschein/draw/sample.py
code served (permissive licence) · get_code("fa5a89b0a62ca754")
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 1, dtype family float, shape (5,) 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)

# Analyze the function:
# - Takes arr as parameter
# - Calls arr.min(), arr.max() on it
# - Does arithmetic operations (subtraction, division)
# - Returns a scaled array
# The function works with array-like objects that support min(), max(), and arithmetic

# Create a small numpy array for testing
arr = np.array([1.0, 2.0, 3.0, 4.0, 5.0])

ARGS = {
    "arr": arr
}
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 ndarray, ndim 1, dtype family float, shape (5,) — shape (5,) 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
str2bool Ran
ran, violated the contract we drafted
this paper's copy was not recorded; identical code first harvested from thuml/LogME
pointer only · get_code("25c7475539e39da4")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

unit_prefix Ran
ran; our contract draft was wrong, not the code
suhoy901/DRAW_pytorch/utils.py
pointer only (licence: NONE) · get_code("070e898b844d8b61")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, suhoy901/DRAW_pytorch when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

xrecons_grid Ran
ran; our contract draft was wrong, not the code
ericjang/draw/plot_data.py
code served (permissive licence) · get_code("a240a4e67c3c2f4a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned ndarray, ndim 2, dtype family float, shape (16, 14) 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

# The function expects:
# - X: (batch_size x img_size) where img_size = B * A
# - batch_size must be a perfect square
# - B: height of each image
# - A: width of each image

# Choose batch_size = 4 (which is 2x2, a perfect square)
# Choose B = 6 (height)
# Choose A = 5 (width)

batch_size = 4
B = 6
A = 5
img_size = B * A

# Create X with shape (batch_size, img_size)
np.random.seed(0)
X = np.random.rand(batch_size, img_size)

ARGS = {
    'X': X,
    'B': B,
    'A': A
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
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 2, dtype family float, shape (16, 14) — shape (16, 14) 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
xrecons_grid Ran
ran on a synthesized input
MoustafaMeshry/draw/plot_data.py
code served (permissive licence) · get_code("99b9ef9642c40ab7")
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 2, dtype family float, shape (10, 14) 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

# The function takes X (reconstructed images), B (height of each image), and A (width of each image)
# X should be (batch_size x img_size) where batch_size is a perfect square
# The function reshapes X into (N, N, B, A) where N = sqrt(batch_size)
# So batch_size must be a perfect square, and img_size must equal B * A

# Set up parameters:
# - batch_size = 4 (perfect square: 2x2)
# - B = 3 (height of each small image)
# - A = 5 (width of each small image)
# - img_size = B * A = 15

batch_size = 4
B = 3
A = 5
img_size = B * A

# Create X with shape (batch_size, img_size)
# X should contain pixel values that will be reshaped into images
X = np.random.rand(batch_size, img_size).astype(np.float32)

ARGS = {
    "X": X,
    "B": B,
    "A": A
}
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 2, dtype family float, shape (10, 14) — shape (10, 14) 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
add_distortions Not yet run
unverified (harvested, no recorded run)
jlindsey15/DRAM/DRAMVAEGAN/DRAMLearnedMetric.py
pointer only (licence: NONE) · get_code("adc0d1230b0b39b0")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jlindsey15/DRAM when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

convertTranslated Not yet run
unverified (harvested, no recorded run)
jlindsey15/DRAM/DRAMVAEGAN/DRAMLearnedMetric.py
pointer only (licence: NONE) · get_code("18f407725809c356")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jlindsey15/DRAM when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

create_sample Not yet run
unverified (harvested, no recorded run)
jlindsey15/DRAM/DRAMVAEGAN/DRAMLearnedMetric.py
pointer only (licence: NONE) · get_code("0942eff650073076")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jlindsey15/DRAM when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

filter_response2histogram Not yet run
unverified (harvested, no recorded run)
MoustafaMeshry/draw/texture_loss.py
code served (permissive licence) · get_code("1b185012743ad260")
How it ran

no invocation recorded.

filterbank Not yet run
unverified (harvested, no recorded run)
MoustafaMeshry/draw/draw.py
code served (permissive licence) · get_code("0108c9c4f32a89d6")
How it ran

no invocation recorded.

get_data Not yet run
unverified (harvested, no recorded run)
singhsidhukuldeep/Generating-Devanagari-Using-DRAW/dataloader.py
code served (permissive licence) · get_code("05b9fee23508018c")
How it ran

no invocation recorded.

im2filter_response Not yet run
unverified (harvested, no recorded run)
MoustafaMeshry/draw/texture_loss.py
code served (permissive licence) · get_code("025b5b9515188170")
How it ran

no invocation recorded.

img_grid Not yet run
unverified (harvested, no recorded run)
jbornschein/draw/sample.py
code served (permissive licence) · get_code("52d20b4814d29ddf")
How it ran

no invocation recorded.

my_batched_dot Not yet run
unverified (harvested, no recorded run)
jbornschein/draw/draw/attention.py
code served (permissive licence) · get_code("c39a03ce39200cdf")
How it ran

no invocation recorded.

Also present in other papers' code

Of 18 samples lifted from this paper's linked repositories, 18 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 3 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 3 later) and 14 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
str2bool 23 other papers
1 dated earlier, 22 later
1410.5401 (2014-10)
Neural Turing Machines
1504.06852 (2015-04)
FlowNet: Learning Optical Flow with Convolutional Networks
1608.00530 (2016-08)
Early Methods for Detecting Adversarial Images
and 20 more not named here
find_model 1 other paper
1 later
1802.04208 (2018-02)
Adversarial Audio Synthesis
load_config 1 other paper
1 later
1802.04208 (2018-02)
Adversarial Audio Synthesis

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

This paper introduces the Deep Recurrent Attentive Writer (DRAW) neural network architecture for image generation. DRAW networks combine a novel spatial attention mechanism that mimics the foveation of the human eye, with a sequential variational auto-encoding framework that allows for the iterative construction of complex images. The system substantially improves on the state of the art for generative models on MNIST, and, when trained on the Street View House Numbers dataset, it generates images that cannot be distinguished from real data with the naked eye.

For agents

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

get_harvested_code_for_paper("1502.04623")
get_code_for_paper("1502.04623")
have("1502.04623")

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

curl https://syntology.ai/api/ran/1502.04623.json

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

[![Syntology run record](https://syntology.ai/api/ran/1502.04623.svg)](https://syntology.ai/paper/1502.04623)

Connect an agent — have() is free.