SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1506.02640 · CVPR · 2015

You Only Look Once: Unified, Real-Time Object Detection

Ali Farhadi, Ross Girshick, Joseph Redmon, Santosh Divvala

arXiv · PDF · Open in the Atlas

Code that ran

89 of 148 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (27 of them constructed an object rather than computing a result; 4 honoured the contract we drafted, 4 violated it and 50 have no contract verdict; 31 failures of our instrument; 59 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/1506.02640.json · badge: /api/ran/1506.02640.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
Banus/caffe-demo community
Papers with Code archive
9 ran · 3 no recorded run
GiaKhangLuu/YOLOv1_from_scratch community
Papers with Code archive
7 ran · 1 no recorded run
DevBruce/YOLOv1-TF2 community (classifier label)
classifier basis not recorded on the link
7 ran · 1 no recorded run
yakhyo/YOLOv1-pt community
not in the paper's declared repositories
7 ran · 0 no recorded run
chihyanghsu0805/object_detection_yolo community
not in the paper's declared repositories
6 ran · 2 no recorded run
westerndigitalcorporation/YOLOv3-in-PyTorch community
not in the paper's declared repositories
5 ran · 3 no recorded run
copy not recorded community
not in the paper's declared repositories
5 ran · 2 no recorded run
3epochs/you-only-look-once community
Papers with Code archive
5 ran · 0 no recorded run
jalotra/Queue-Detection community
not in the paper's declared repositories
4 ran · 1 no recorded run
DavianYang/yolo.ai community
Papers with Code archive
3 ran · 5 no recorded run
vvasilo/yolov3_pytorch_ros community
not in the paper's declared repositories
3 ran · 3 no recorded run
1991yuyang/YOLOV1-PYTORCH community
Papers with Code archive
3 ran · 0 no recorded run
abajaj945/Ship-Detection-using-Tensorflow community
not in the paper's declared repositories
2 ran · 4 no recorded run
nsoul97/yolov1_pytorch community
not in the paper's declared repositories
2 ran · 1 no recorded run
zjZSTU/YOLO_v1 community
not in the paper's declared repositories
2 ran · 0 no recorded run
easternRainy/MSDS-DL-Final-YOLO-Implement community
not in the paper's declared repositories
2 ran · 0 no recorded run
shkim960520/YOLO-v1-for-studying community
not in the paper's declared repositories
2 ran · 0 no recorded run
s1r2ikanth/Atunomous-Driving-Application--For-Car-detection community
not in the paper's declared repositories
1 ran · 6 no recorded run
patrick013/object-detection---yolov3 community
not in the paper's declared repositories
1 ran · 1 no recorded run
leon-liangwu/MaskYolo_Caffe community
not in the paper's declared repositories
1 ran · 1 no recorded run
Kartik-Aggarwal/Real-Time-Traffic-Sign-Detection community
Papers with Code archive
1 ran · 0 no recorded run
bian0505/Pad_Me community
not in the paper's declared repositories
1 ran · 0 no recorded run
euske/miniyolo community
not in the paper's declared repositories
1 ran · 0 no recorded run
ivanwhaf/yolov1-pytorch community
not in the paper's declared repositories
1 ran · 0 no recorded run
tanjeffreyz/yolo-v1 community
not in the paper's declared repositories
1 ran · 0 no recorded run
FunctionStatusWhere it lives
BACKBONE Ran
ran on a synthesized input · constructed an object; no result was computed
yakhyo/YOLOv1-pt/nets/nn.py
code served (permissive licence) · get_code("12df2a4d4e2164d9")
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 an instance of its own class, BACKBONE: the run constructed the object and did not call it, so no result was computed. 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 = {
    "num_classes": 10,
    "init_weight": True,
}
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 BACKBONE — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a BACKBONE 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
BasicConv2d Ran
ran on a synthesized input
zjZSTU/YOLO_v1/py/lib/models/yolo_v1.py
code served (permissive licence) · get_code("9c798969164d0e3f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 4, dtype family float, shape (2, 4, 5, 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 torch.nn as nn

class BasicConv2d(nn.Module):
    """
    结合BN的卷积操作
    """

    def __init__(self, in_channels, out_channels, kernel_size=1, stride=1, padding=0, norm_layer=None, relu_layer=None):
        super(BasicConv2d, self).__init__()
        if norm_layer is None:
            norm_layer = nn.BatchNorm2d
        if relu_layer is None:
            relu_layer = nn.RReLU
        self.conv = nn.Conv2d(in_channels, out_channels, kernel_size, stride=stride, padding=padding)
        self.bn = norm_layer(out_channels, eps=0.001)
        self.relu = relu_layer(lower=0.1, upper=0.1, inplace=True)

    def forward(self, x):
        x = self.conv(x)
        x = self.bn(x)
        return self.relu(x)


INIT_ARGS = {
    'in_channels': 3,
    'out_channels': 4,
    'kernel_size': 3,
    'stride': 1,
    'padding': 1,
    'norm_layer': None,
    'relu_layer': None,
}

ARGS = {
    'x': torch.randn(2, 3, 5, 5),
}

if __name__ == '__main__':
    model = BasicConv2d(**INIT_ARGS)
    model.eval()
    output = model(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Expected shape: torch.Size([2, 4, 5, 5])")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 4, dtype family float, shape (2, 4, 5, 5) — shape (2, 4, 5, 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
BottleNeckBlock Ran
ran on a synthesized input · constructed an object; no result was computed
GiaKhangLuu/YOLOv1_from_scratch/model/models.py
pointer only (licence: NONE) · get_code("145089cb92dfe192")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, GiaKhangLuu/YOLOv1_from_scratch 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.

CNNBlock Ran
ran on a synthesized input · constructed an object; no result was computed
easternRainy/MSDS-DL-Final-YOLO-Implement/yolo.py
pointer only (licence: NONE) · get_code("6f5fcdaa1a66346b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, easternRainy/MSDS-DL-Final-YOLO-Implement 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.

CNNBlock Ran
ran on a synthesized input
shkim960520/YOLO-v1-for-studying/model.py
pointer only (licence: NONE) · get_code("925c307bbd79f4c9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shkim960520/YOLO-v1-for-studying 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.

Conv Ran
ran on a synthesized input · constructed an object; no result was computed
yakhyo/YOLOv1-pt/nets/nn.py
code served (permissive licence) · get_code("e51b597949497410")
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 an instance of its own class, Conv: the run constructed the object and did not call it, so no result was computed. 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 = {
    'c1': 3,
    'c2': 16,
    'k': 3,
    's': 1,
    'p': None,
    'd': 1,
    'g': 1,
    'act': True
}
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 Conv — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Conv 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
Conv1X1 Ran
ran on a synthesized input · constructed an object; no result was computed
1991yuyang/YOLOV1-PYTORCH/model.py
pointer only (licence: NONE) · get_code("6650a36a34670d48")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, 1991yuyang/YOLOV1-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.

Conv3X3 Ran
ran on a synthesized input · constructed an object; no result was computed
1991yuyang/YOLOV1-PYTORCH/model.py
pointer only (licence: NONE) · get_code("2a16f500c758e0fb")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, 1991yuyang/YOLOV1-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.

ConvBlock Ran
ran on a synthesized input · constructed an object; no result was computed
DavianYang/yolo.ai/yolo/models/arch/yolov1.py
code served (permissive licence) · get_code("0a87cc68078b8130")
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 an instance of its own class, ConvBlock: the run constructed the object and did not call it, so no result was computed. 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
from torch import nn
from torch import Tensor

torch.manual_seed(0)

# Define ARGS dictionary with all ConvBlock parameters
ARGS = {
    'in_channels': 3,
    'out_channels': 16,
    'kernel_size': 3,
    'stride': 1,
    'padding': 1,
    'dilation': 1,
    'groups': 1,
    'bias': False,
    'activation': 'relu'
}
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 ConvBlock — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a ConvBlock 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
ConvLayer Ran
ran on a synthesized input
westerndigitalcorporation/YOLOv3-in-PyTorch/src/model.py
pointer only (licence: NOASSERTION) · get_code("a2da5672bbc086d3")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, westerndigitalcorporation/YOLOv3-in-PyTorch (NOASSERTION). 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.

ConvModule Ran
ran on a synthesized input
nsoul97/yolov1_pytorch/code/model.py
pointer only (licence: NONE) · get_code("2390515679b99339")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nsoul97/yolov1_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.

ConvWithBatchNorm Ran
ran on a synthesized input · constructed an object; no result was computed
GiaKhangLuu/YOLOv1_from_scratch/model/models.py
pointer only (licence: NONE) · get_code("3ca60117f4fea257")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, GiaKhangLuu/YOLOv1_from_scratch 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.

ConvexHull Ran
ran on a synthesized input · constructed an object; no result was computed
jalotra/Queue-Detection/src/yolo/detector.py
pointer only (licence: NONE) · get_code("8d9039359a33a5fe")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jalotra/Queue-Detection 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.

DarkNet53BackBone Ran
ran on a synthesized input
westerndigitalcorporation/YOLOv3-in-PyTorch/src/model.py
pointer only (licence: NOASSERTION) · get_code("49d95f317c8411ea")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, westerndigitalcorporation/YOLOv3-in-PyTorch (NOASSERTION). 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.

Darknet Ran
ran on a synthesized input · constructed an object; no result was computed
3epochs/you-only-look-once/darknet.py
pointer only (licence: NONE) · get_code("161d0afc6942ee8a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, 3epochs/you-only-look-once 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.

DarknetConv2D Ran
ran; our contract draft was wrong, not the code
s1r2ikanth/Atunomous-Driving-Application--For-Car-detection/yad2k/models/keras_yolo.py
pointer only (licence: NOASSERTION) · get_code("d6893ebc8516e8db")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, s1r2ikanth/Atunomous-Driving-Application--For-Car-detection (NOASSERTION). 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.

Detect Ran
ran on a synthesized input · constructed an object; no result was computed
Kartik-Aggarwal/Real-Time-Traffic-Sign-Detection/Codes/models/yolo.py
pointer only (licence: NONE) · get_code("95b9ab17a245bb24")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Kartik-Aggarwal/Real-Time-Traffic-Sign-Detection 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.

DetectionLayer Ran
ran on a synthesized input · constructed an object; no result was computed
3epochs/you-only-look-once/darknet.py
pointer only (licence: NONE) · get_code("da55cf9866cb41e1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, 3epochs/you-only-look-once 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.

EmptyLayer Ran
ran on a synthesized input · constructed an object; no result was computed
3epochs/you-only-look-once/darknet.py
pointer only (licence: NONE) · get_code("c3b8937867a41dde")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, 3epochs/you-only-look-once 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.

Flatten Ran
ran on a synthesized input · constructed an object; no result was computed
yakhyo/YOLOv1-pt/nets/nn.py
code served (permissive licence) · get_code("b445acdcf8d90d37")
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 an instance of its own class, Flatten: the run constructed the object and did not call it, so no result was computed. 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)

class Flatten(nn.Module):
    def __init__(self):
        super(Flatten, self).__init__()

    @staticmethod
    def forward(x):
        return x.view(x.size(0), -1)

# Create an instance of Flatten
flatten_module = Flatten()

# Create a test input tensor with shape (batch_size=2, channels=4, height=3, width=5)
# This is a 4D tensor typical for image data
x = torch.randn(2, 4, 3, 5)

# The Flatten module doesn't take any __init__ parameters, so ARGS is empty
ARGS = {}
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 Flatten — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Flatten 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
GlobalAvgPool2d Ran
ran on a synthesized input
yakhyo/YOLOv1-pt/nets/nn.py
code served (permissive licence) · get_code("32281a62359fa309")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 2, 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 torch
import torch.nn as nn

class GlobalAvgPool2d(nn.Module):

    def __init__(self):
        super(GlobalAvgPool2d, self).__init__()

    @staticmethod
    def forward(x):
        return torch.mean(x.view(x.size(0), x.size(1), -1), dim=2)


INIT_ARGS = {}

ARGS = {
    'x': torch.randn(2, 4, 3, 5)
}

if __name__ == '__main__':
    model = GlobalAvgPool2d(**INIT_ARGS)
    output = model(**ARGS)
    print(output.shape)
    print(output)
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 2, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
HEAD Ran
ran on a synthesized input · constructed an object; no result was computed
yakhyo/YOLOv1-pt/nets/nn.py
code served (permissive licence) · get_code("2353b99f0ed68149")
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 an instance of its own class, HEAD: the run constructed the object and did not call it, so no result was computed. 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)

# Define the helper functions and classes that HEAD depends on
def pad(k, p):
    if p is None:
        p = k // 2
    return p

class Conv(nn.Module):
    def __init__(self, c1, c2, k, s=1, p=None, d=1, g=1, act=True):
        super(Conv, self).__init__()
        self.conv = nn.Conv2d(c1, c2, k, s, pad(k, p), dilation=d, groups=g, bias=False)
        self.bn = nn.BatchNorm2d(c2, momentum=0.03, eps=0.001)
        self.act = nn.LeakyReLU(0.01, inplace=True) if act else nn.ReLU(inplace=True)

    def forward(self, x):
        return self.act(self.bn(self.conv(x)))

class Flatten(nn.Module):
    def __init__(self):
        super(Flatten, self).__init__()

    @staticmethod
    def forward(x):
        return x.view(x.size(0), -1)

class HEAD(nn.Module):
    def __init__(self, fs, nb, nc):
        super(HEAD, self).__init__()
        self.conv = nn.Sequential(Conv(1024, 1024, 3), Conv(1024, 1024, 3, 2), Conv(1024, 1024, 3), Conv(1024, 1024, 3))
        self.detect = nn.Sequential(Flatten(), nn.Linear(7 * 7 * 1024, 4096), nn.LeakyReLU(0.01, inplace=True), nn.Dropout(0.5), nn.Linear(4096, fs * fs * (5 * nb + nc)), nn.Sigmoid())

    def forward(self, x):
        x = self.conv(x)
        x = self.detect(x)
        return x

# Setup arguments for HEAD
fs = 7  # grid size
nb = 2  # number of bounding boxes
nc = 5  # number of classes

ARGS = {
    'fs': fs,
    'nb': nb,
    'nc': nc
}
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 HEAD — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a HEAD 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
LocallyConnected2d Ran
ran on a synthesized input
nsoul97/yolov1_pytorch/code/model.py
pointer only (licence: NONE) · get_code("0a3ef0b1d1f175b9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nsoul97/yolov1_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.

ORIYOLO Ran
ran on a synthesized input · constructed an object; no result was computed
1991yuyang/YOLOV1-PYTORCH/model.py
pointer only (licence: NONE) · get_code("d77878590fc2562c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, 1991yuyang/YOLOV1-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.

Point Ran
ran on a synthesized input · constructed an object; no result was computed
jalotra/Queue-Detection/src/yolo/detector.py
pointer only (licence: NONE) · get_code("28f729bdf2cd5201")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jalotra/Queue-Detection 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.

Region_Mask Ran
ran on a synthesized input
bian0505/Pad_Me/modelclass.py
code served (permissive licence) · get_code("e2e6a6ad8e928647")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 4, 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 torch
import torch.nn as nn
import torch.nn.functional as F

class Region_Mask(nn.Module):

    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(3, 64, 7)
        self.bt1 = nn.BatchNorm2d(64)
        self.conv2 = nn.Conv2d(64, 192, 3)
        self.bt2 = nn.BatchNorm2d(192)
        self.conv30 = nn.Conv2d(192, 128, 1)
        self.conv31 = nn.Conv2d(128, 256, 3)
        self.conv32 = nn.Conv2d(256, 256, 1)
        self.conv33 = nn.Conv2d(256, 512, 3)
        self.bt3 = nn.BatchNorm2d(512)
        self.conv400 = nn.Conv2d(512, 256, 1)
        self.conv410 = nn.Conv2d(256, 512, 3)
        self.conv401 = nn.Conv2d(512, 256, 1)
        self.conv411 = nn.Conv2d(256, 512, 3)
        self.conv402 = nn.Conv2d(512, 256, 1)
        self.conv412 = nn.Conv2d(256, 512, 3)
        self.conv403 = nn.Conv2d(512, 256, 1)
        self.conv413 = nn.Conv2d(256, 512, 3)
        self.conv42 = nn.Conv2d(512, 512, 1)
        self.conv43 = nn.Conv2d(512, 1024, 3)
        self.bt4 = nn.BatchNorm2d(1024)
        self.conv500 = nn.Conv2d(1024, 512, 1)
        self.conv510 = nn.Conv2d(512, 1024, 3)
        self.conv501 = nn.Conv2d(1024, 512, 1)
        self.conv511 = nn.Conv2d(512, 1024, 3)
        self.conv52 = nn.Conv2d(1024, 1024, 3)
        self.bt5 = nn.BatchNorm2d(1024)
        self.dp5 = nn.Dropout(0.2)
        self.conv60 = nn.Conv2d(1024, 1024, 3)
        self.conv61 = nn.Conv2d(1024, 1024, 3)
        self.bt6 = nn.BatchNorm2d(1024)
        self.dp6 = nn.Dropout(0.2)
        self.fc1 = nn.Linear(4096, 4096)
        self.fc2 = nn.Linear(4096, 3125)

    def forward(self, x):
        out = self.conv1(x)
        out = self.bt1(out)
        out = F.relu(out)
        out = F.max_pool2d(out, 2, 2)
        out = self.conv2(out)
        out = self.bt2(out)
        out = F.relu(out)
        out = F.max_pool2d(out, 2, 2)
        out = self.conv30(out)
        out = self.conv31(out)
        out = self.conv32(out)
        out = self.conv33(out)
        out = self.bt3(out)
        out = F.relu(out)
        out = F.max_pool2d(out, 2, 2)
        out = self.conv400(out)
        out = self.conv410(out)
        out = self.conv401(out)
        out = self.conv411(out)
        out = self.conv402(out)
        out = self.conv412(out)
        out = self.conv403(out)
        out = self.conv413(out)
        out = self.conv42(out)
        out = self.conv43(out)
        out = self.bt4(out)
        out = F.relu(out)
        out = self.conv500(out)
        out = self.conv510(out)
        out = self.conv501(out)
        out = self.conv511(out)
        out = self.conv52(out)
        out = self.bt5(out)
        out = self.dp5(out)
        out = F.relu(out)
        out = F.max_pool2d(out, 2, 2)
        out = self.conv60(out)
        out = self.conv61(out)
        out = self.bt6(out)
        out = self.dp6(out)
        out = F.relu(out)
        out = out.view(out.size(0), -1)
        out = self.fc1(out)
        out = self.fc2(out)
        out_final = out.view([-1, 25, 25, 5])
        return out_final.float()

INIT_ARGS = {}

ARGS = {
    'x': torch.randn(2, 3, 256, 256)
}

if __name__ == '__main__':
    model = Region_Mask(**INIT_ARGS)
    model.eval()
    output = model(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Output dtype: {output.dtype}")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 4, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
ResBlock Ran
ran on a synthesized input · constructed an object; no result was computed
DavianYang/yolo.ai/yolo/models/arch/yolov1.py
code served (permissive licence) · get_code("ffa921481b6a54c3")
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 an instance of its own class, ResBlock: the run constructed the object and did not call it, so no result was computed. 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
from inspect import isfunction

torch.manual_seed(0)

class MockBlockConfig:
    def __init__(self, filters, kernel_size, stride, padding):
        self.filters = filters
        self.kernel_size = kernel_size
        self.stride = stride
        self.padding = padding

def get_activation_layer(activation):
    assert activation is not None
    if isfunction(activation):
        return activation()
    elif isinstance(activation, str):
        if activation == 'relu':
            return nn.ReLU(inplace=True)
        elif activation == 'leakyrelu':
            return nn.LeakyReLU(0.1, inplace=True)
        elif activation == 'silu':
            return nn.SiLU()
        else:
            raise NotImplementedError()
    else:
        assert isinstance(activation, nn.Module)
        return activation

class ConvBlock(nn.Module):
    def __init__(self, in_channels: int, out_channels: int, kernel_size, stride=1, padding=0, dilation=1, groups=1, bias=False, activation=lambda: nn.ReLU(inplace=True)):
        super().__init__()
        self.activate = activation is not None
        self.conv = nn.Conv2d(in_channels, out_channels, kernel_size, stride, padding, dilation, groups, bias)
        self.bn = nn.BatchNorm2d(out_channels)
        if self.activate:
            self.act = get_activation_layer(activation)

    def forward(self, x):
        x = self.bn(self.conv(x))
        if self.activate:
            x = self.act(x)
        return x

# Create mock block configurations
block1 = MockBlockConfig(filters=8, kernel_size=3, stride=1, padding=1)
block2 = MockBlockConfig(filters=8, kernel_size=3, stride=1, padding=1)

blocks = [block1, block2]

ARGS = {
    "channels": 8,
    "blocks": blocks
}
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 ResBlock — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a ResBlock 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
ResBlock Ran
ran on a synthesized input
westerndigitalcorporation/YOLOv3-in-PyTorch/src/model.py
pointer only (licence: NOASSERTION) · get_code("f10b817692f95152")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, westerndigitalcorporation/YOLOv3-in-PyTorch (NOASSERTION). 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.

YOLOLayer Ran
ran on a synthesized input
vvasilo/yolov3_pytorch_ros/src/yolov3_pytorch_ros/models/models.py
code served (permissive licence) · get_code("8255067215f41bdd")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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 torch
import torch.nn as nn
import numpy as np

# INIT_ARGS for YOLOLayer.__init__(anchors, num_classes, img_dim)
INIT_ARGS = {
    'anchors': [(10, 13), (16, 30), (33, 23)],  # 3 anchors
    'num_classes': 5,
    'img_dim': 416
}

# ARGS for YOLOLayer.forward(x, targets=None)
# x should have shape (batch, num_anchors * bbox_attrs, grid, grid)
# bbox_attrs = 5 + num_classes = 5 + 5 = 10
# num_anchors = 3
# So channel dimension = 3 * 10 = 30

batch_size = 2
num_anchors = 3
num_classes = 5
bbox_attrs = 5 + num_classes  # 10
grid_size = 13
img_dim = 416

# Input tensor x: (batch, num_anchors * bbox_attrs, grid, grid)
x_input = torch.randn(batch_size, num_anchors * bbox_attrs, grid_size, grid_size)

# Targets: (batch, max_objects, 5) where each row is [class_id, x_center, y_center, width, height]
# All coordinates are normalized (0-1)
max_objects = 3
targets_input = torch.zeros(batch_size, max_objects, 5)

# First batch, first object
targets_input[0, 0, :] = torch.tensor([2.0, 0.5, 0.5, 0.3, 0.4])  # class 2, centered
# First batch, second object
targets_input[0, 1, :] = torch.tensor([1.0, 0.3, 0.7, 0.2, 0.3])  # class 1

# Second batch, first object
targets_input[1, 0, :] = torch.tensor([4.0, 0.6, 0.4, 0.25, 0.35])  # class 4

ARGS = {
    'x': x_input,
    'targets': targets_input
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
YOLONet Ran
ran on a synthesized input · constructed an object; no result was computed
euske/miniyolo/model.py
code served (permissive licence) · get_code("d04bf9a09a736a52")
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 an instance of its own class, YOLONet: the run constructed the object and did not call it, so no result was computed. 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)

device = torch.device('cpu')
nvals = 30

ARGS = {
    'device': device,
    'nvals': nvals,
}
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 YOLONet — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a YOLONet 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
YOLO_v1 Ran
ran on a synthesized input
zjZSTU/YOLO_v1/py/lib/models/yolo_v1.py
code served (permissive licence) · get_code("56cda48b112cd414")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 3, dtype family float, shape (2, 49, 30) 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 torch.nn as nn

class BasicConv2d(nn.Module):
    """
    结合BN的卷积操作
    """

    def __init__(self, in_channels, out_channels, kernel_size=1, stride=1, padding=0, norm_layer=None, relu_layer=None):
        super(BasicConv2d, self).__init__()
        if norm_layer is None:
            norm_layer = nn.BatchNorm2d
        if relu_layer is None:
            relu_layer = nn.RReLU
        self.conv = nn.Conv2d(in_channels, out_channels, kernel_size, stride=stride, padding=padding)
        self.bn = norm_layer(out_channels, eps=0.001)
        self.relu = relu_layer(lower=0.1, upper=0.1, inplace=True)

    def forward(self, x):
        x = self.conv(x)
        x = self.bn(x)
        return self.relu(x)

class YOLO_v1(nn.Module):

    def __init__(self, S=7, B=2, C=20):
        super(YOLO_v1, self).__init__()
        conv_block = BasicConv2d
        self.features = nn.Sequential(conv_block(3, 64, kernel_size=7, stride=2), nn.MaxPool2d(2, stride=2), conv_block(64, 192, kernel_size=3, padding=1), nn.MaxPool2d(2, stride=2), conv_block(192, 128, kernel_size=1), conv_block(128, 256, kernel_size=3, padding=1), conv_block(256, 256, kernel_size=1), conv_block(256, 512, kernel_size=3, padding=1), nn.MaxPool2d(2, stride=2), conv_block(512, 256, kernel_size=1), conv_block(256, 512, kernel_size=3, padding=1), conv_block(512, 256, kernel_size=1), conv_block(256, 512, kernel_size=3, padding=1), conv_block(512, 256, kernel_size=1), conv_block(256, 512, kernel_size=3, padding=1), conv_block(512, 256, kernel_size=1), conv_block(256, 512, kernel_size=3, padding=1), conv_block(512, 512, kernel_size=1), conv_block(512, 1024, kernel_size=3, padding=1), nn.MaxPool2d(2, stride=2), conv_block(1024, 512, kernel_size=1), conv_block(512, 1024, kernel_size=3, padding=1), conv_block(1024, 512, kernel_size=1), conv_block(512, 1024, kernel_size=3, padding=1), conv_block(1024, 1024, kernel_size=3, padding=1), conv_block(1024, 1024, kernel_size=3, stride=2), conv_block(1024, 1024, kernel_size=3, padding=1), conv_block(1024, 1024, kernel_size=3, padding=1))
        self.avgpool = nn.AdaptiveAvgPool2d((7, 7))
        self.classifier = nn.Sequential(nn.Linear(7 * 7 * 1024, 4096), nn.ReLU(inplace=True), nn.Linear(4096, S * S * (B * 5 + C)))
        self.S = S
        self.B = B
        self.C = C

    def forward(self, x):
        """
        :param x: [N, C, H, W]
        :return:
        """
        assert len(x.shape) == 4
        N = x.shape[0]
        x = self.features(x)
        x = self.avgpool(x)
        x = torch.flatten(x, 1)
        x = self.classifier(x)
        x = torch.sigmoid(x)
        return x.reshape(N, self.S * self.S, self.B * 5 + self.C)


INIT_ARGS = {
    'S': 7,
    'B': 2,
    'C': 20
}

ARGS = {
    'x': torch.randn(2, 3, 448, 448)
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 3, dtype family float, shape (2, 49, 30) — shape (2, 49, 30) 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
YOLOv1 Ran
ran on a synthesized input · constructed an object; no result was computed
yakhyo/YOLOv1-pt/nets/nn.py
code served (permissive licence) · get_code("1eb112cfe550ab4e")
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 an instance of its own class, YOLOv1: the run constructed the object and did not call it, so no result was computed. 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 = {
    'fs': 7,
    'nb': 2,
    'nc': 20,
    'pretrained_backbone': False,
}
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 YOLOv1 — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a YOLOv1 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
YOLOv1 Ran
ran on a synthesized input · constructed an object; no result was computed
ivanwhaf/yolov1-pytorch/models/yolov1.py
code served (permissive licence) · get_code("81ea9a33a5ab4d93")
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 an instance of its own class, YOLOv1: the run constructed the object and did not call it, so no result was computed. 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)

# YOLOv1 takes three parameters: S, B, num_classes
# S = grid size (typically 7)
# B = number of bounding boxes per grid cell (typically 2)
# num_classes = number of object classes (typically 20 for PASCAL VOC)

ARGS = {
    "S": 7,
    "B": 2,
    "num_classes": 20,
}
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 YOLOv1 — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a YOLOv1 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
YOLOv1 Ran
ran on a synthesized input · constructed an object; no result was computed
tanjeffreyz/yolo-v1/models.py
pointer only (licence: NONE) · get_code("320df1877e192cc4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, tanjeffreyz/yolo-v1 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.

YOLOv3 Ran
ran on a synthesized input · constructed an object; no result was computed
MasoumehVahedi/Yolo-Object-Detection/model.py
pointer only (licence: NONE) · get_code("cd696cd247a1d09d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, MasoumehVahedi/Yolo-Object-Detection 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.

Yolo Ran
ran on a synthesized input · constructed an object; no result was computed
easternRainy/MSDS-DL-Final-YOLO-Implement/yolo.py
pointer only (licence: NONE) · get_code("176e983db06d554b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, easternRainy/MSDS-DL-Final-YOLO-Implement 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.

YoloBackbone Ran
ran on a synthesized input · constructed an object; no result was computed
GiaKhangLuu/YOLOv1_from_scratch/model/models.py
pointer only (licence: NONE) · get_code("04b987f5b68bd18e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, GiaKhangLuu/YOLOv1_from_scratch 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.

YoloLayer Ran
ran on a synthesized input
westerndigitalcorporation/YOLOv3-in-PyTorch/src/model.py
pointer only (licence: NOASSERTION) · get_code("4bd7300fab880607")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, westerndigitalcorporation/YOLOv3-in-PyTorch (NOASSERTION). 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.

YoloOutput Ran
ran on a synthesized input · constructed an object; no result was computed
GiaKhangLuu/YOLOv1_from_scratch/model/models.py
pointer only (licence: NONE) · get_code("bb3b7889b05c820a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, GiaKhangLuu/YOLOv1_from_scratch 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.

Yolov1 Ran
ran on a synthesized input · constructed an object; no result was computed
Abdulrahman-Adel/Object-Detection/YOLOv1/model.py
code served (permissive licence) · get_code("df07a80fb38aace5")
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 an instance of its own class, Yolov1: the run constructed the object and did not call it, so no result was computed. 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,
    "split_size": 7,
    "num_boxes": 2,
    "num_classes": 20,
}
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 Yolov1 — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Yolov1 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
Yolov1 Ran
ran on a synthesized input
shkim960520/YOLO-v1-for-studying/model.py
pointer only (licence: NONE) · get_code("08b9a36f02f7601f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shkim960520/YOLO-v1-for-studying 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.

_count_conv_weights Ran
ran, honoured the contract we drafted
abajaj945/Ship-Detection-using-Tensorflow/model_layers.py
code served (permissive licence) · get_code("056c8fee6ae1bef5")
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 int, ndim 0, dtype family int, 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
import torch.nn as nn

torch.manual_seed(0)

# Create mock tensor objects with get_shape() method that returns mock shape objects
class MockShape:
    def __init__(self, dims):
        self.dims = dims
    
    def __getitem__(self, idx):
        return self.dims[idx]

class MockTensor:
    def __init__(self, shape_dims):
        self.shape_dims = shape_dims
    
    def get_shape(self):
        return MockShape(self.shape_dims)

# Create input tensor with shape [batch, height, width, channels_in]
# Using distinct sizes: 2 (batch), 4 (height), 5 (width), 3 (input channels)
input_tensor = MockTensor([2, 4, 5, 3])

# Create output tensor with shape [batch, height, width, channels_out]
# Using distinct sizes: 2 (batch), 4 (height), 5 (width), 7 (output channels)
output_tensor = MockTensor([2, 4, 5, 7])

# Filter size (kernel size)
filters_size = 3

ARGS = {
    "input": input_tensor,
    "output": output_tensor,
    "filters_size": filters_size
}
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 int, ndim 0, dtype family int, 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
_layer_stats Ran
ran; our contract draft was wrong, not the code
abajaj945/Ship-Detection-using-Tensorflow/model_layers.py
code served (permissive licence) · get_code("c40601ebebe70c7c")
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 dict. 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)

# Create a mock tensor with get_shape() method
# The function calls output.get_shape()[1], [2], [3]
# So we need a tensor-like object with a get_shape() method
class MockTensor:
    def __init__(self, shape):
        self.shape = shape
    
    def get_shape(self):
        return self

    def __getitem__(self, idx):
        return self.shape[idx]

# Construct arguments
info = {'layers': 0, 'weights': 0, 'description': ''}
layer_name = 'conv1'
output = MockTensor((2, 32, 32, 64))  # (batch, height, width, depth)
layers_incr = 1
weights_incr = 9408

ARGS = {
    'info': info,
    'layer_name': layer_name,
    'output': output,
    'layers_incr': layers_incr,
    'weights_incr': weights_incr
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
bbox_iou Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
vvasilo/yolov3_pytorch_ros/src/yolov3_pytorch_ros/models/models.py
code served (permissive licence) · get_code("359c906d4db69e92")
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) and returned Tensor, ndim 1, dtype family float, shape (2,) 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 two small bounding box tensors
# Format: [x1, y1, x2, y2] for x1y1x2y2=True format
# Shape: (batch_size, 4)
box1 = torch.tensor([
    [10.0, 20.0, 50.0, 60.0],
    [15.0, 25.0, 55.0, 65.0]
], dtype=torch.float32)

box2 = torch.tensor([
    [30.0, 40.0, 70.0, 80.0],
    [20.0, 30.0, 60.0, 70.0]
], dtype=torch.float32)

ARGS = {
    "box1": box1,
    "box2": box2,
    "x1y1x2y2": True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 1, dtype family float, shape (2,) — shape (2,) 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
bbox_selector Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Ugenteraan/YOLOv1/YOLO v1.py
pointer only (licence: NONE) · get_code("14e1a3b5a738b277")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Ugenteraan/YOLOv1 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.

build_targets Ran
ran; our contract draft was wrong, not the code
vvasilo/yolov3_pytorch_ros/src/yolov3_pytorch_ros/models/models.py
code served (permissive licence) · get_code("892cb24caebb0a10")
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 math

torch.manual_seed(0)

# Build minimal tensors matching function usage
batch_size = 2
num_anchors = 3
grid_size = 4
num_classes = 5
max_targets = 3
img_dim = 416

# pred_boxes: [batch, num_anchors, grid_size, grid_size, 4]
pred_boxes = torch.randn(batch_size, num_anchors, grid_size, grid_size, 4)

# pred_conf: [batch, num_anchors, grid_size, grid_size]
pred_conf = torch.rand(batch_size, num_anchors, grid_size, grid_size)

# pred_cls: [batch, num_anchors, grid_size, grid_size, num_classes]
pred_cls = torch.randn(batch_size, num_anchors, grid_size, grid_size, num_classes)

# target: [batch, max_targets, 5] where each target is [class_id, x, y, w, h] (normalized)
target = torch.zeros(batch_size, max_targets, 5)
target[0, 0] = torch.tensor([1.0, 0.5, 0.5, 0.3, 0.3])
target[0, 1] = torch.tensor([2.0, 0.2, 0.7, 0.2, 0.25])
target[1, 0] = torch.tensor([3.0, 0.8, 0.3, 0.25, 0.2])

# anchors: list of (width, height) tuples
anchors = [(0.5, 0.5), (1.0, 1.0), (2.0, 2.0)]

# Scalars
num_anchors_param = 3
num_classes_param = 5
grid_size_param = 4
ignore_thres = 0.5
img_dim_param = 416

ARGS = {
    'pred_boxes': pred_boxes,
    'pred_conf': pred_conf,
    'pred_cls': pred_cls,
    'target': target,
    'anchors': anchors,
    'num_anchors': num_anchors_param,
    'num_classes': num_classes_param,
    'grid_size': grid_size_param,
    'ignore_thres': ignore_thres,
    'img_dim': img_dim_param,
}
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 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
calc_iou Ran
ran on a synthesized input
DevBruce/YOLOv1-TF2/libs/iou.py
code served (permissive licence) · get_code("b63e002a7056cc61")
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 EagerTensor, ndim 3, dtype family float, shape (2, 2, 2) 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 tensorflow as tf
import numpy as np

class YOLOConfig:
    def __init__(self):
        self.input_width = 416
        self.input_height = 416
        self.eps = 1e-6

# Create config
cfg = YOLOConfig()

# Create pred_boxes: [cell_size, cell_size, boxes_per_cell, 5]
# Using small dimensions: 2x3x2x5
cell_size = 2
boxes_per_cell = 2
pred_boxes = tf.constant(
    np.random.rand(cell_size, cell_size, boxes_per_cell, 5).astype(np.float32) * 200,
    dtype=tf.float32
)
# Make it valid: ensure xmin < xmax, ymin < ymax
pred_boxes_np = pred_boxes.numpy()
pred_boxes_np[:, :, :, 0] = 50  # xmin
pred_boxes_np[:, :, :, 1] = 60  # ymin
pred_boxes_np[:, :, :, 2] = 150  # xmax
pred_boxes_np[:, :, :, 3] = 160  # ymax
pred_boxes_np[:, :, :, 4] = 0.9  # confidence
pred_boxes = tf.constant(pred_boxes_np, dtype=tf.float32)

# Create gt_box: [4] with relative coordinates [x_center_rel, y_center_rel, width_rel, height_rel]
gt_box = tf.constant(
    [0.3, 0.4, 0.2, 0.25],
    dtype=tf.float32
)

ARGS = {
    "pred_boxes": pred_boxes,
    "gt_box": gt_box,
    "cfg": cfg,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 3, dtype family float, shape (2, 2, 2) — shape (2, 2, 2) 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
calculate_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kiranrgupta26/ObjectDetection/WheatHeadDetection.py
pointer only (licence: NONE) · get_code("10cfbec4f3be1ace")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, kiranrgupta26/ObjectDetection 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.

clockwise Ran
ran, violated the contract we drafted
jalotra/Queue-Detection/src/yolo/detector.py
pointer only (licence: NONE) · get_code("1c48ff2f9d302176")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jalotra/Queue-Detection 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.

compute_iou Ran
ran on a synthesized input
Everina/car-detection-yolo/yolo_v1.py
code served (permissive licence) · get_code("84b11777c5a839c6")
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 EagerTensor, ndim 2, dtype family float, shape (2, 2) 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 tensorflow as tf
import numpy as np

# Set seed for reproducibility
tf.random.set_seed(0)

# Create sample box tensors
# boxes format: [center_x, center_y, width, height]
# Shape: (batch_size, num_boxes, 4)

boxes1 = tf.constant([
    [[10.0, 10.0, 4.0, 6.0], [20.0, 20.0, 8.0, 10.0]],
    [[5.0, 5.0, 2.0, 3.0], [15.0, 15.0, 4.0, 5.0]]
], dtype=tf.float32)

boxes2 = tf.constant([
    [[12.0, 11.0, 5.0, 7.0], [19.0, 21.0, 6.0, 8.0]],
    [[6.0, 6.0, 3.0, 4.0], [14.0, 16.0, 5.0, 6.0]]
], dtype=tf.float32)

ARGS = {
    "boxes1": boxes1,
    "boxes2": boxes2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 2, dtype family float, shape (2, 2) — shape (2, 2) 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
conv_layer Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
chihyanghsu0805/object_detection_yolo/yolo_v3.py
pointer only (licence: NONE) · get_code("ddf34c500d70f342")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, chihyanghsu0805/object_detection_yolo 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.

convert_cellbox_to_corner_bbox Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
GiaKhangLuu/YOLOv1_from_scratch/model/models.py
pointer only (licence: NONE) · get_code("7e493fee89015615")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, GiaKhangLuu/YOLOv1_from_scratch 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.

convert_cellbox_to_xywh Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
GiaKhangLuu/YOLOv1_from_scratch/model/models.py
pointer only (licence: NONE) · get_code("e604e68b98ad3d67")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, GiaKhangLuu/YOLOv1_from_scratch 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.

counter_clockwise Ran
ran, violated the contract we drafted
jalotra/Queue-Detection/src/yolo/detector.py
pointer only (licence: NONE) · get_code("f0af3d51eba51c84")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jalotra/Queue-Detection 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.

cpu_nms Ran
ran; our contract draft was wrong, not the code
chihyanghsu0805/object_detection_yolo/yolo_v3.py
pointer only (licence: NONE) · get_code("8cae105928c8d4d5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, chihyanghsu0805/object_detection_yolo 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.

create_modules Ran
ran; our contract draft was wrong, not the code
3epochs/you-only-look-once/darknet.py
pointer only (licence: NONE) · get_code("1f8a9bd89881ff05")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, 3epochs/you-only-look-once 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.

darknet53 Ran
ran; our contract draft was wrong, not the code
chihyanghsu0805/object_detection_yolo/yolo_v3.py
pointer only (licence: NONE) · get_code("b8f085d6c5d3158c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, chihyanghsu0805/object_detection_yolo 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.

get_activation_layer Ran
ran; our contract draft was wrong, not the code
DavianYang/yolo.ai/yolo/models/arch/yolov1.py
code served (permissive licence) · get_code("3ddd214dfdd86aa1")
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 ReLU. 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
from torch import nn
from inspect import isfunction

torch.manual_seed(0)

# Test case 1: activation as a string
ARGS = {
    'activation': 'relu'
}
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 ReLU — a ReLU output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ReLU 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
get_boxes Ran
ran; our contract draft was wrong, not the code
Banus/caffe-demo/yolo_detection.py
pointer only (licence: NOASSERTION) · get_code("bc3cd3f8c2d09fe5")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, Banus/caffe-demo (NOASSERTION). 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.

get_candidate_objects Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Banus/caffe-demo/yolo_detection.py
pointer only (licence: NOASSERTION) · get_code("6f90433e21ba8f19")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, Banus/caffe-demo (NOASSERTION). 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.

Showing the first 60. An agent gets every one from get_harvested_code_for_paper("1506.02640").

Also present in other papers' code

Of 148 samples lifted from this paper's linked repositories, 148 carry a function identity and 0 do not. Of the 147 distinct functions with an identity, 10 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 10 later) and 137 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
sigmoid
brick of this paper's assembly: helper
12 other papers
12 later · 1 only as an assembly brick
1602.01783 (2016-02)
Asynchronous Methods for Deep Reinforcement Learning
2106.12930 (2021-06)
VinDr-SpineXR: A deep learning framework for spinal lesions detection and classification from radiographs
136730727 (2022 (year only))
Self-Supervised Sparse Representation for Video Anomaly Detection
and 9 more not named here
bbox_iou
brick of this paper's assembly: helper
1 other paper
1 later
1602.00763 (2016-02)
Simple Online and Realtime Tracking
cpu_nms
brick of this paper's assembly: helper
1 other paper
1 later
1902.04103 (2019-02)
Bag of Freebies for Training Object Detection Neural Networks
findboxes
brick of this paper's assembly: entry
1 other paper
1 later
1612.08242 (2016-12)
YOLO9000: Better, Faster, Stronger
process_boxes
brick of this paper's assembly: helper
1 other paper
1 later
1612.08242 (2016-12)
YOLO9000: Better, Faster, Stronger
process_wh
brick of this paper's assembly: helper
1 other paper
1 later
1612.08242 (2016-12)
YOLO9000: Better, Faster, Stronger
process_xy
brick of this paper's assembly: helper
1 other paper
1 later
1612.08242 (2016-12)
YOLO9000: Better, Faster, Stronger
space_to_depth_x2
brick of this paper's assembly: helper
1 other paper
1 later
1612.08242 (2016-12)
YOLO9000: Better, Faster, Stronger
space_to_depth_x2_output_shape
brick of this paper's assembly: helper
1 other paper
1 later
1612.08242 (2016-12)
YOLO9000: Better, Faster, Stronger
upsample
brick of this paper's assembly: helper
1 other paper
1 later
1612.08242 (2016-12)
YOLO9000: Better, Faster, Stronger

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

More are linked than shown here.

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

Abstract

We present YOLO, a new approach to object detection. Prior work on object detection repurposes classifiers to perform detection. Instead, we frame object detection as a regression problem to spatially separated bounding boxes and associated class probabilities. A single neural network predicts bounding boxes and class probabilities directly from full images in one evaluation. Since the whole detection pipeline is a single network, it can be optimized end-to-end directly on detection performance. Our unified architecture is extremely fast. Our base YOLO model processes images in real-time at 45 frames per second. A smaller version of the network, Fast YOLO, processes an astounding 155 frames per second while still achieving double the mAP of other real-time detectors. Compared to state-of-the-art detection systems, YOLO makes more localization errors but is less likely to predict false positives on background. Finally, YOLO learns very general representations of objects. It outperforms other detection methods, including DPM and R-CNN, when generalizing from natural images to other domains like artwork.

For agents

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

get_harvested_code_for_paper("1506.02640")
get_code_for_paper("1506.02640")
have("1506.02640")

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

curl https://syntology.ai/api/ran/1506.02640.json

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

[![Syntology run record](https://syntology.ai/api/ran/1506.02640.svg)](https://syntology.ai/paper/1506.02640)

Connect an agent — have() is free.