SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1905.00641 · 2019

RetinaFace: Single-stage Dense Face Localisation in the Wild

arXiv · PDF · Open in the Atlas

Code that ran

73 of 91 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; 1 honoured the contract we drafted, 4 violated it and 45 have no contract verdict; 23 failures of our instrument; 18 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/1905.00641.json · badge: /api/ran/1905.00641.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
2275519762/RetinaFace_Pytorch community (classifier label)
classifier basis not recorded on the link
9 ran · 1 no recorded run
nhatduy19599/insightface-master community (classifier label)
classifier basis not recorded on the link
9 ran · 0 no recorded run
bubbliiiing/retinaface-keras community (classifier label)
classifier basis not recorded on the link
8 ran · 3 no recorded run
TinDang97/face_recognition community (classifier label)
classifier basis not recorded on the link
7 ran · 2 no recorded run
code-trip/insightface-mxnet community (classifier label)
classifier basis not recorded on the link
6 ran · 0 no recorded run
hukkelas/DSFD-Pytorch-Inference community (classifier label)
classifier basis not recorded on the link
5 ran · 1 no recorded run
sajjjadayobi/FaceLib community (classifier label)
classifier basis not recorded on the link
4 ran · 4 no recorded run
hhj1897/face_detection community (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
StanislasBertrand/RetinaFace-tf2 community (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
serengil/deepface community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
neelabh17/MAVI-Face reimplementation (classifier label)
classifier basis: classifier_2025
2 ran · 1 no recorded run
iChenning/facedetection community (classifier label)
classifier basis not recorded on the link
2 ran · 1 no recorded run
bleakie/RetinaDetection community (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
GuoQuanhao/RetinaFace-Paddle community (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
qiaoxiu/facenetRetinaFace community (classifier label)
classifier basis not recorded on the link
1 ran · 2 no recorded run
RuisongZhou/RetinaFace.pytorch community (classifier label)
classifier basis not recorded on the link
1 ran · 1 no recorded run
yakhyo/retinaface-pytorch reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
peteryuX/retinaface-tf2 reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
yangyucheng000/RetinaFace reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
prajinkhadka/face_det_check reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
vladimirwest/insightface_cinematic community (classifier label)
classifier basis not recorded on the link
0 ran · 1 no recorded run
discipleofhamilton/RetinaFace community (classifier label)
classifier basis not recorded on the link
0 ran · 1 no recorded run
FunctionStatusWhere it lives
BboxHead Ran
ran on a synthesized input
bubbliiiing/retinaface-keras/nets/retinaface.py
code served (permissive licence) · get_code("3dd288c9fd68d9ad")
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 KerasTensor. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
from tensorflow import keras
from keras.layers import Input
import tensorflow as tf

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

# Create a small input tensor with shape (batch_size, height, width, channels)
# Using distinct dimensions: batch=2, height=4, width=3, channels=5
input_tensor = keras.Input(shape=(4, 3, 5))

# Define the arguments dictionary
ARGS = {
    "inputs": input_tensor,
    "num_anchors": 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
keras, numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type KerasTensor — a KerasTensor 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
ClassHead Ran
ran on a synthesized input
bubbliiiing/retinaface-keras/nets/retinaface.py
code served (permissive licence) · get_code("62174cec5466dc02")
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 KerasTensor. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
from tensorflow import keras
from tensorflow.keras.layers import Input
from tensorflow.keras.backend import variable

# Create a small input tensor
# The function expects inputs to be a tensor that can be passed through Conv2D
# Conv2D expects a 4D tensor: (batch, height, width, channels)
batch_size = 2
height = 4
width = 3
channels = 5

# Create a mock input using Keras Input layer
input_tensor = Input(shape=(height, width, channels), batch_size=batch_size)

# Define ARGS dict with all parameters
ARGS = {
    "inputs": input_tensor,
    "num_anchors": 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
keras, numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type KerasTensor — a KerasTensor 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
IOU Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
bleakie/RetinaDetection/rcnn/core/tester.py
code served (permissive licence) · get_code("9348919b3a8be935")
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 float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# The function expects two bounding boxes in format [x1, y1, x2, y2]
# where (x1, y1) is top-left corner and (x2, y2) is bottom-right corner

# Create two overlapping bounding boxes
Reframe = [10.0, 20.0, 50.0, 60.0]  # Box 1: top-left (10, 20), bottom-right (50, 60)
GTframe = [30.0, 40.0, 70.0, 80.0]  # Box 2: top-left (30, 40), bottom-right (70, 80)

ARGS = {
    'Reframe': Reframe,
    'GTframe': GTframe
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
MobileNet Ran
ran on a synthesized input
bubbliiiing/retinaface-keras/nets/mobilenet.py
code served (permissive licence) · get_code("897011a8b72fc39f")
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 keras
from keras import backend as K
from keras.layers import Input

# Set seeds for reproducibility
import numpy as np
np.random.seed(0)

# Create a small input tensor for the MobileNet function
# The function expects a Keras Input or tensor
# img_input should be a 4D tensor: (batch, height, width, channels)
# Using batch=2, height=8, width=8, channels=3 (small sizes)
img_input = Input(shape=(8, 8, 3), batch_size=2)

ARGS = {
    'img_input': img_input,
    'depth_multiplier': 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
keras, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
ResNet50 Ran
ran on a synthesized input
bubbliiiing/retinaface-keras/nets/resnet.py
code served (permissive licence) · get_code("3655d96a3170ca28")
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.

from keras import Input
import numpy as np

# Create a small input tensor for ResNet50
# The function expects a Keras Input or tensor
# ResNet50 processes images, so we need shape (batch, height, width, channels)
# Using batch=2, height=32, width=32, channels=3 for small tensors
inputs = Input(shape=(32, 32, 3))

ARGS = {
    'inputs': inputs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
keras, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
anchors_plane Ran
ran on a synthesized input
code-trip/insightface-mxnet/python-package/insightface/model_zoo/face_detection.py
code served (permissive licence) · get_code("15c53f53a69a6879")
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 4, dtype family float, shape (4, 5, 3, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Construct valid arguments for anchors_plane
height = 4
width = 5
stride = 16
base_anchors = np.array([
    [-8, -8, 8, 8],
    [-16, -16, 16, 16],
    [-32, -32, 32, 32]
], dtype=np.float32)

ARGS = {
    "height": height,
    "width": width,
    "stride": stride,
    "base_anchors": base_anchors
}
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 4, dtype family float, shape (4, 5, 3, 4) — shape (4, 5, 3, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
box_smooth_l1 Ran
ran on a synthesized input
bubbliiiing/retinaface-keras/nets/retinaface_training.py
code served (permissive licence) · get_code("fc93aaaf4087913f")
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 function. 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 keras
import tensorflow as tf
import numpy as np

def box_smooth_l1(sigma=1, weights=1):
    sigma_squared = sigma ** 2

    def _smooth_l1(y_true, y_pred):
        regression = y_pred
        regression_target = y_true[:, :, :-1]
        anchor_state = y_true[:, :, -1]
        indices = tf.where(keras.backend.equal(anchor_state, 1))
        regression = tf.gather_nd(regression, indices)
        regression_target = tf.gather_nd(regression_target, indices)
        regression_diff = regression - regression_target
        regression_diff = keras.backend.abs(regression_diff)
        regression_loss = tf.where(keras.backend.less(regression_diff, 1.0 / sigma_squared), 0.5 * sigma_squared * keras.backend.pow(regression_diff, 2), regression_diff - 0.5 / sigma_squared)
        normalizer = keras.backend.maximum(1, keras.backend.shape(indices)[0])
        normalizer = keras.backend.cast(normalizer, dtype=keras.backend.floatx())
        loss = keras.backend.sum(regression_loss) / normalizer
        return loss * weights
    return _smooth_l1

# Create y_true: shape (batch_size, num_anchors, num_values)
# Last dimension contains regression targets (:-1) and anchor_state (-1)
# We need at least one anchor with state=1 for the function to work
batch_size = 2
num_anchors = 3
num_regression_dims = 4

y_true_np = np.random.randn(batch_size, num_anchors, num_regression_dims + 1).astype(np.float32)
# Set anchor_state to 1 for some anchors so indices is not empty
y_true_np[:, :, -1] = 1.0

y_true = tf.constant(y_true_np)

# Create y_pred: shape (batch_size, num_anchors, num_regression_dims)
y_pred_np = np.random.randn(batch_size, num_anchors, num_regression_dims).astype(np.float32)
y_pred = tf.constant(y_pred_np)

ARGS = {
    "sigma": 1.0,
    "weights": 1.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
keras, numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
build_from_cfg Ran
ran on a synthesized input
hukkelas/DSFD-Pytorch-Inference/face_detection/registry.py
code served (permissive licence) · get_code("8241f65f228cbc9d")
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 SimpleModule. 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


# Create a simple Registry class to mimic the expected registry
class Registry:
    def __init__(self, name):
        self.name = name
        self._module_dict = {}
    
    def register(self, name, module_class):
        self._module_dict[name] = module_class
        return module_class
    
    def get(self, name):
        return self._module_dict.get(name)


# Create a simple test module
class SimpleModule(nn.Module):
    def __init__(self, in_features=4, out_features=3, **kwargs):
        super().__init__()
        self.linear = nn.Linear(in_features, out_features)
    
    def forward(self, x):
        return self.linear(x)


# Create a registry and register the module
test_registry = Registry('test_registry')
test_registry.register('SimpleModule', SimpleModule)

# Build the configuration dict
cfg = {
    'type': 'SimpleModule',
    'in_features': 8,
    'out_features': 5
}

ARGS = {
    'cfg': cfg,
    'registry': test_registry
}
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 SimpleModule — a SimpleModule output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleModule 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
check_keys Ran
ran, violated the contract we drafted
neelabh17/MAVI-Face/test_widerface.py
code served (permissive licence) · get_code("1d429d5224fe49ff")
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 bool, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a simple model
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 8)
        self.linear2 = nn.Linear(8, 3)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

# Instantiate the model
model = SimpleModel()

# Create a pretrained state dict with some overlapping and some extra keys
pretrained_state_dict = {
    'linear1.weight': torch.randn(8, 4),
    'linear1.bias': torch.randn(8),
    'linear2.weight': torch.randn(3, 8),
    'linear2.bias': torch.randn(3),
    'extra_layer.weight': torch.randn(2, 2),  # unused key
}

ARGS = {
    'model': model,
    'pretrained_state_dict': pretrained_state_dict,
}
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 bool, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
conf_loss Ran
ran on a synthesized input
bubbliiiing/retinaface-keras/nets/retinaface_training.py
code served (permissive licence) · get_code("f4473f7b1b6d52e7")
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 function. 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 tensorflow as tf
import numpy as np

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

# Create small tensors for testing
batch_size = 2
num_boxes = 4
num_classes = 3

# y_true: shape (batch_size, num_boxes, num_classes + 1)
# Last dimension: [class_0, class_1, class_2, is_positive]
y_true_data = np.array([
    [
        [1.0, 0.0, 0.0, 1.0],  # positive box, class 0
        [0.0, 1.0, 0.0, 1.0],  # positive box, class 1
        [0.0, 0.0, 0.0, 0.0],  # negative box
        [0.0, 0.0, 0.0, 0.0],  # negative box
    ],
    [
        [0.0, 1.0, 0.0, 1.0],  # positive box, class 1
        [0.0, 0.0, 0.0, 0.0],  # negative box
        [0.0, 0.0, 0.0, 0.0],  # negative box
        [0.0, 0.0, 0.0, 0.0],  # negative box
    ]
], dtype=np.float32)
y_true = tf.constant(y_true_data)

# y_pred: shape (batch_size, num_boxes, num_classes)
# Predicted class probabilities
y_pred_data = np.array([
    [
        [0.8, 0.1, 0.1],
        [0.1, 0.8, 0.1],
        [0.3, 0.3, 0.4],
        [0.2, 0.2, 0.6],
    ],
    [
        [0.1, 0.7, 0.2],
        [0.4, 0.3, 0.3],
        [0.3, 0.3, 0.4],
        [0.25, 0.25, 0.5],
    ]
], dtype=np.float32)
y_pred = tf.constant(y_pred_data)

ARGS = {
    "neg_pos_ratio": 7,
    "negatives_for_hard": 100,
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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; our contract draft was wrong, not the code
sajjjadayobi/FaceLib/facelib/FacialExpression/models/resnet.py
code served (permissive licence) · get_code("d9def42110729a85")
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 Conv2d. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# The function conv1x1 takes in_planes, out_planes, and stride as parameters
# It returns an nn.Conv2d layer configured for 1x1 convolution
# We need to provide valid integer arguments for these parameters

ARGS = {
    "in_planes": 3,
    "out_planes": 64,
    "stride": 1
}
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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
conv2d_dw_group Ran
ran on a synthesized input
TinDang97/face_recognition/models/rpn.py
code served (permissive licence) · get_code("1db2dd8d77f384e4")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 4, dtype family float, shape (2, 3, 3, 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
import torch.nn.functional as functional

torch.manual_seed(0)

# x needs to be a 4D tensor with shape (batch, channel, height, width)
# The function reshapes it to (1, batch*channel, height, width)
# So we need: batch, channel, height, width to be distinct and meaningful
x = torch.randn(2, 3, 5, 4)  # (batch=2, channel=3, height=5, width=4)

# kernel needs to be a 4D tensor with shape (batch, channel, kernel_h, kernel_w)
# The function extracts batch and channel from kernel.shape[:2]
# Then reshapes to (batch*channel, 1, kernel_h, kernel_w)
kernel = torch.randn(2, 3, 3, 3)  # (batch=2, channel=3, kernel_h=3, kernel_w=3)

ARGS = {
    "x": x,
    "kernel": kernel,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 3, 3, 2) — shape (2, 3, 3, 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
conv3x3 Ran
ran; our contract draft was wrong, not the code
sajjjadayobi/FaceLib/facelib/FacialExpression/models/resnet.py
code served (permissive licence) · get_code("160bb14bd76201b4")
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 Conv2d. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "in_planes": 3,
    "out_planes": 64,
    "stride": 1,
    "groups": 1,
    "dilation": 1,
}
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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
conv3x3 Ran
ran; our contract draft was wrong, not the code
TinDang97/face_recognition/models/resnet.py
pointer only (licence: MIT); the stored copy get_code serves, from lixiaotong97/DSU, is pointer only (licence: NONE) · get_code("fac5364e2f53c6db")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from lixiaotong97/DSU, and GitHub detected no licence in that repository 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_block Ran
ran on a synthesized input
bubbliiiing/retinaface-keras/nets/resnet.py
code served (permissive licence) · get_code("b4b2f767b1cfc5ed")
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 KerasTensor. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
from tensorflow import keras
from keras import layers
from keras.layers import Input

# Create a small input tensor with shape (batch, height, width, channels)
# Using batch=2, height=4, width=4, channels=3 for a tiny but valid tensor
input_tensor = Input(shape=(4, 4, 3))

# Define arguments for conv_block function
ARGS = {
    'input_tensor': input_tensor,
    'kernel_size': (3, 3),
    'filters': (64, 64, 256),  # filters1, filters2, filters3
    'stage': 2,
    'block': 'a',
    'strides': (2, 2)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
keras, numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type KerasTensor — a KerasTensor 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
conv_bn Ran
ran; our contract draft was wrong, not the code
2275519762/RetinaFace_Pytorch/models/net.py
code served (permissive licence) · get_code("f5414e78ff7a815f")
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 Sequential. 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)

# The function conv_bn takes inp, oup, stride, and leaky parameters
# inp: input channels (int)
# oup: output channels (int)
# stride: stride for Conv2d (int, default=1)
# leaky: negative_slope for LeakyReLU (float, default=0)

ARGS = {
    'inp': 3,
    'oup': 16,
    'stride': 1,
    'leaky': 0.1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
conv_bn Ran
ran on a synthesized input
TinDang97/face_recognition/models/face_detector.py
code served (permissive licence) · get_code("ad3b8a6dd472d83c")
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 Sequential. 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 = {
    'inp': 3,
    'oup': 16,
    'stride': 1,
    'leaky': 0.1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
conv_bn1X1 Ran
ran; our contract draft was wrong, not the code
2275519762/RetinaFace_Pytorch/models/net.py
code served (permissive licence) · get_code("bb07aa213e6aacc9")
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 Sequential. 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 = {
    'inp': 3,
    'oup': 16,
    'stride': 1,
    'leaky': 0.1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
conv_bn_no_relu Ran
ran; our contract draft was wrong, not the code
2275519762/RetinaFace_Pytorch/models/net.py
code served (permissive licence) · get_code("aacb8f654f54d534")
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 Sequential. 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 = {
    "inp": 3,
    "oup": 16,
    "stride": 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
conv_bn_no_relu Ran
ran on a synthesized input
RuisongZhou/RetinaFace.pytorch/models/net.py
code served (permissive licence) · get_code("3dfbecba8d9d1dba")
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 Sequential. 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 = {
    "inp": 4,
    "oup": 8,
    "stride": 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
create_map Ran
ran on a synthesized input
nhatduy19599/insightface-master/deploy/mxnet_to_ort.py
code served (permissive licence) · get_code("b20d28b3252ea1cb")
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

class GraphMember:
    def __init__(self, name):
        self.name = name

torch.manual_seed(0)

# Create a list of graph members with distinct names
graph_member_list = [
    GraphMember("node_1"),
    GraphMember("node_2"),
    GraphMember("node_3"),
]

ARGS = {
    "graph_member_list": graph_member_list,
}
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
decode Ran
ran on a synthesized input
hhj1897/face_detection/ibug/face_detection/s3fd/utils.py
code served (permissive licence) · get_code("957481035acfb48d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# loc: tensor of shape [num_priors, 4] containing location predictions
loc = torch.randn(2, 4)

# priors: tensor of shape [num_priors, 4] containing prior boxes in center-offset form
priors = torch.randn(2, 4)

# variances: list of floats
variances = [0.1, 0.2]

ARGS = {
    "loc": loc,
    "priors": priors,
    "variances": variances,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
decode_boxes Ran
ran on a synthesized input
bubbliiiing/retinaface-keras/vision_for_anchors.py
code served (permissive licence) · get_code("5fb249a30ff39c2e")
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 (2, 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

# Create small test data for the decode_boxes function
# The function expects:
# - mbox_loc: shape (N, 4) - bounding box location predictions
# - mbox_ldm: shape (N, 10) - landmark predictions (5 landmarks x 2 coordinates)
# - mbox_priorbox: shape (N, 4) - prior boxes in format [xmin, ymin, xmax, ymax]

np.random.seed(0)

# Number of boxes
N = 2

# mbox_loc: (2, 4) - location offsets for center_x, center_y, width, height
mbox_loc = np.random.randn(N, 4).astype(np.float32) * 0.5

# mbox_ldm: (2, 10) - 5 landmarks with x, y coordinates each
mbox_ldm = np.random.randn(N, 10).astype(np.float32) * 0.1

# mbox_priorbox: (2, 4) - prior boxes in format [xmin, ymin, xmax, ymax]
# Create boxes with reasonable coordinates (0-1 range, typical for normalized coords)
mbox_priorbox = np.array([
    [0.1, 0.1, 0.5, 0.5],
    [0.2, 0.2, 0.7, 0.8]
], dtype=np.float32)

ARGS = {
    'mbox_loc': mbox_loc,
    'mbox_ldm': mbox_ldm,
    'mbox_priorbox': mbox_priorbox
}
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 (2, 14) — shape (2, 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
densenet121 Ran
ran; our contract draft was wrong, not the code
sajjjadayobi/FaceLib/facelib/FacialExpression/models/densenet.py
code served (permissive licence) · get_code("933ea7e525b5c882")
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 DenseNet. 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)

# Build a minimal DenseNet121 model
def densenet121(pretrained=False, progress=True, **kwargs):
    r"""Densenet-121 model from
    `"Densely Connected Convolutional Networks" <https://arxiv.org/pdf/1608.06993.pdf>`_

    Args:
        pretrained (bool): If True, returns a model pre-trained on ImageNet
        progress (bool): If True, displays a progress bar of the download to stderr
    """
    model = DenseNet(num_init_features=64, growth_rate=32, block_config=(6, 12, 24, 16),
                     **kwargs)
    return model

# Minimal implementations needed for the function to work
class _DenseLayer(nn.Sequential):
    def __init__(self, num_input_features, growth_rate, bn_size, drop_rate, memory_efficient=False):
        super(_DenseLayer, self).__init__()
        self.add_module('norm1', nn.BatchNorm2d(num_input_features))
        self.add_module('relu1', nn.ReLU(inplace=True))
        self.add_module('conv1', nn.Conv2d(num_input_features, bn_size * growth_rate, kernel_size=1, stride=1, bias=False))
        self.add_module('norm2', nn.BatchNorm2d(bn_size * growth_rate))
        self.add_module('relu2', nn.ReLU(inplace=True))
        self.add_module('conv2', nn.Conv2d(bn_size * growth_rate, growth_rate, kernel_size=3, stride=1, padding=1, bias=False))
        self.drop_rate = drop_rate
        self.memory_efficient = memory_efficient

    def forward(self, *prev_features):
        return prev_features[0]

class _DenseBlock(nn.Module):
    def __init__(self, num_layers, num_input_features, bn_size, growth_rate, drop_rate, memory_efficient=False):
        super(_DenseBlock, self).__init__()
        for i in range(num_layers):
            layer = _DenseLayer(num_input_features + i * growth_rate, growth_rate=growth_rate, bn_size=bn_size, drop_rate=drop_rate, memory_efficient=memory_efficient)
            self.add_module('denselayer%d' % (i + 1), layer)

    def forward(self, init_features):
        return init_features

class _Transition(nn.Sequential):
    def __init__(self, num_input_features, num_output_features):
        super(_Transition, self).__init__()
        self.add_module('norm', nn.BatchNorm2d(num_input_features))
        self.add_module('relu', nn.ReLU(inplace=True))
        self.add_module('conv', nn.Conv2d(num_input_features, num_output_features, kernel_size=1, stride=1, bias=False))
        self.add_module('pool', nn.AvgPool2d(kernel_size=2, stride=2))

class DenseNet(nn.Module):
    def __init__(self, growth_rate=32, block_config=(6, 12, 24, 16), num_init_features=64, bn_size=4, drop_rate=0, memory_efficient=False):
        super(DenseNet, self).__init__()
        self.features = nn.Sequential()
        self.growth_rate = growth_rate
        self.block_config = block_config
        self.num_init_features = num_init_features
        self.bn_size = bn_size
        self.drop_rate = drop_rate
        self.memory_efficient = memory_efficient

    def forward(self, x):
        return x

ARGS = {
    'pretrained': False,
    'progress': 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 DenseNet — a DenseNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DenseNet 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
distance2bbox Ran
ran on a synthesized input
nhatduy19599/insightface-master/python-package/insightface/model_zoo/scrfd.py
code served (permissive licence) · get_code("1d23600a36442d09")
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 (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Create points tensor: shape (n, 2) where n=2, representing [x, y] coordinates
points = torch.tensor([
    [10.0, 20.0],
    [50.0, 60.0]
], dtype=torch.float32)

# Create distance tensor: shape (n, 4) representing [left, top, right, bottom] distances
distance = torch.tensor([
    [5.0, 10.0, 15.0, 20.0],
    [8.0, 12.0, 18.0, 25.0]
], dtype=torch.float32)

# Define max_shape as image dimensions (height, width)
max_shape = (100, 120)

ARGS = {
    "points": points,
    "distance": distance,
    "max_shape": max_shape
}
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 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
distance2kps Ran
ran on a synthesized input
nhatduy19599/insightface-master/python-package/insightface/model_zoo/scrfd.py
code served (permissive licence) · get_code("e8f5d8d07e1e560b")
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 (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Construct points tensor: shape (n, 2) where n=2 points, each with [x, y]
points = torch.tensor([
    [10.0, 20.0],
    [30.0, 40.0]
], dtype=torch.float32)

# Construct distance tensor: shape (n, 4) representing distances to 4 boundaries
# (left, top, right, bottom) for each point
distance = torch.tensor([
    [5.0, 3.0, 7.0, 2.0],
    [4.0, 6.0, 8.0, 1.0]
], dtype=torch.float32)

# max_shape is a tuple representing (height, width) of the image
max_shape = (100, 150)

ARGS = {
    "points": points,
    "distance": distance,
    "max_shape": max_shape
}
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 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
find_onnx_file Ran
ran on a synthesized input
nhatduy19599/insightface-master/python-package/insightface/model_zoo/model_zoo.py
code served (permissive licence) · get_code("fc633d5737805a00")
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 str. 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 os
import tempfile
import shutil

# Create a temporary directory with a test .onnx file
temp_dir = tempfile.mkdtemp()
test_onnx_file = os.path.join(temp_dir, "model.onnx")
with open(test_onnx_file, 'w') as f:
    f.write("dummy onnx content")

ARGS = {
    "dir_path": temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
find_params_file Ran
ran on a synthesized input
code-trip/insightface-mxnet/python-package/insightface/model_zoo/model_store.py
code served (permissive licence) · get_code("8e01f2544321014d")
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 str. 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 os
import tempfile
import shutil

# Create a temporary directory with some .params files
temp_dir = tempfile.mkdtemp()
test_file_1 = os.path.join(temp_dir, "model_step_100.params")
test_file_2 = os.path.join(temp_dir, "model_step_200.params")
test_file_3 = os.path.join(temp_dir, "model_step_150.params")

# Create the files
open(test_file_1, 'w').close()
open(test_file_2, 'w').close()
open(test_file_3, 'w').close()

ARGS = {
    "dir_path": temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
fixed_image_standardization Ran
ran, violated the contract we drafted
qiaoxiu/facenetRetinaFace/face/infer2.py
code served (permissive licence) · get_code("f3274d8580b03de4")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 4, dtype family float, shape (2, 4, 3, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function takes an image_tensor and performs standardization
# image_tensor should be a torch.Tensor with numeric values
# The function subtracts 127.5 and divides by 128.0

image_tensor = torch.randint(0, 256, (2, 4, 3, 3), dtype=torch.float32)

ARGS = {
    "image_tensor": image_tensor
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 4, 3, 3) — shape (2, 4, 3, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
flip_data Ran
ran on a synthesized input
nhatduy19599/insightface-master/recognition/oneflow_face/insightface_val.py
code served (permissive licence) · get_code("5b7430811591e445")
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 4, dtype family float, shape (2, 4, 3, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# The function flip_data takes images as input
# It flips along axis=2 and converts to float32
# We need to create a numpy array with at least 3 dimensions

# Create a small image array: batch_size=2, height=4, width=3, channels=3
# Using distinct sizes for clarity
images = np.random.randint(0, 256, size=(2, 4, 3, 3), dtype=np.uint8)

ARGS = {
    "images": images
}
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 4, dtype family float, shape (2, 4, 3, 3) — shape (2, 4, 3, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
generate_anchors Ran
ran on a synthesized input
code-trip/insightface-mxnet/python-package/insightface/model_zoo/face_detection.py
code served (permissive licence) · get_code("4b292d55fb7ca5d5")
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 (9, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

ARGS = {
    'base_size': 16,
    'ratios': np.array([0.5, 1, 2]),
    'scales': 2 ** np.arange(3, 6),
    'stride': 16
}
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 (9, 4) — shape (9, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
generate_anchors_fpn Ran
ran on a synthesized input
code-trip/insightface-mxnet/python-package/insightface/model_zoo/face_detection.py
code served (permissive licence) · get_code("470f084357a89d01")
How it ran

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

import numpy as np

# Construct a configuration dictionary for generate_anchors_fpn
# The function expects a dict where keys are stride values (as strings or ints)
# and values are dicts containing 'BASE_SIZE', 'RATIOS', and 'SCALES'

cfg = {
    '32': {
        'BASE_SIZE': 32,
        'RATIOS': [0.5, 1, 2],
        'SCALES': [8, 16, 32]
    },
    '16': {
        'BASE_SIZE': 16,
        'RATIOS': [0.5, 1, 2],
        'SCALES': [4, 8, 16]
    },
    '8': {
        'BASE_SIZE': 8,
        'RATIOS': [0.5, 1],
        'SCALES': [2, 4]
    }
}

ARGS = {
    'cfg': cfg
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list — a list 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
get_block Ran
ran on a synthesized input
TinDang97/face_recognition/models/model_irse.py
code served (permissive licence) · get_code("5700283c6de360ea")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 2, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
from collections import namedtuple

torch.manual_seed(0)

class Bottleneck(namedtuple('Block', ['in_channel', 'depth', 'stride'])):
    """A named tuple describing a ResNet block."""
    pass

def get_block(in_channel, depth, num_units, stride=2):
    return [Bottleneck(in_channel, depth, stride)] + [Bottleneck(depth, depth, 1) for _ in range(num_units - 1)]

ARGS = {
    'in_channel': 3,
    'depth': 64,
    'num_units': 4,
    'stride': 2
}
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 list, ndim 2, dtype family int — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_blocks Ran
ran on a synthesized input
TinDang97/face_recognition/models/model_irse.py
code served (permissive licence) · get_code("3262857e15d46904")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list. 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
from collections import namedtuple

torch.manual_seed(0)

class Bottleneck(namedtuple('Block', ['in_channel', 'depth', 'stride'])):
    """A named tuple describing a ResNet block."""

def get_block(in_channel, depth, num_units, stride=2):
    return [Bottleneck(in_channel, depth, stride)] + [Bottleneck(depth, depth, 1) for _ in range(num_units - 1)]

def get_blocks(num_layers):
    if num_layers == 50:
        blocks = [get_block(in_channel=64, depth=64, num_units=3), get_block(in_channel=64, depth=128, num_units=4), get_block(in_channel=128, depth=256, num_units=14), get_block(in_channel=256, depth=512, num_units=3)]
    elif num_layers == 100:
        blocks = [get_block(in_channel=64, depth=64, num_units=3), get_block(in_channel=64, depth=128, num_units=13), get_block(in_channel=128, depth=256, num_units=30), get_block(in_channel=256, depth=512, num_units=3)]
    else:
        blocks = [get_block(in_channel=64, depth=64, num_units=3), get_block(in_channel=64, depth=128, num_units=8), get_block(in_channel=128, depth=256, num_units=36), get_block(in_channel=256, depth=512, num_units=3)]
    return blocks

ARGS = {
    "num_layers": 50
}
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 list — a list 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
get_gt_boxes Ran
ran; our contract draft was wrong, not the code
2275519762/RetinaFace_Pytorch/widerface_evaluate/evaluation.py
code served (permissive licence) · get_code("b8864cd681dd62de")
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 os
import tempfile
import numpy as np
from scipy.io import savemat

# Create a temporary directory with mock .mat files
temp_dir = tempfile.mkdtemp()

# Create mock data for the .mat files
facebox_list = np.array([[1, 2, 3, 4], [5, 6, 7, 8]], dtype=object)
event_list = np.array([['event1'], ['event2']], dtype=object)
file_list = np.array([['file1.jpg'], ['file2.jpg']], dtype=object)
hard_gt_list = np.array([[10, 20, 30, 40]], dtype=object)
medium_gt_list = np.array([[50, 60, 70, 80]], dtype=object)
easy_gt_list = np.array([[90, 100, 110, 120]], dtype=object)

# Save the mock data to .mat files
savemat(os.path.join(temp_dir, 'wider_face_val.mat'), {
    'face_bbx_list': facebox_list,
    'event_list': event_list,
    'file_list': file_list
})

savemat(os.path.join(temp_dir, 'wider_hard_val.mat'), {
    'gt_list': hard_gt_list
})

savemat(os.path.join(temp_dir, 'wider_medium_val.mat'), {
    'gt_list': medium_gt_list
})

savemat(os.path.join(temp_dir, 'wider_easy_val.mat'), {
    'gt_list': easy_gt_list
})

ARGS = {
    'gt_dir': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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
get_gt_boxes_from_txt Ran
ran; our contract draft was wrong, not the code
2275519762/RetinaFace_Pytorch/widerface_evaluate/evaluation.py
code served (permissive licence) · get_code("fa5b759a22a26bb3")
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 os
import tempfile
import numpy as np

# Create a temporary directory for cache
temp_dir = tempfile.mkdtemp()

# Create a temporary gt_path file with valid ground truth box data
gt_file_path = os.path.join(temp_dir, 'gt_boxes.txt')

# Write sample ground truth data in the expected format
gt_content = """--image_001
info_line
10.5 20.3 50.7 60.2
15.0 25.0 55.0 65.0
--image_002
info_line
30.1 40.2 70.3 80.4
35.5 45.5 75.5 85.5
--image_003
info_line
5.0 10.0 100.0 110.0
"""

with open(gt_file_path, 'w') as f:
    f.write(gt_content)

ARGS = {
    'gt_path': gt_file_path,
    'cache_dir': temp_dir
}
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 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
get_model_file Ran
ran; our contract draft was wrong, not the code
nhatduy19599/insightface-master/python-package/insightface/model_zoo/model_store.py
code served (permissive licence) · get_code("36d98138ea733ae9")
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 str. 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 os
import tempfile
import shutil

# Create a temporary directory for testing
temp_dir = tempfile.mkdtemp()

# Create a mock model directory structure with a .params file
model_name = 'arcface_r100_v1'
model_root = os.path.join(temp_dir, '.insightface', 'models')
model_dir = os.path.join(model_root, model_name)
os.makedirs(model_dir, exist_ok=True)

# Create a dummy .params file
params_file = os.path.join(model_dir, 'model-0000.params')
with open(params_file, 'w') as f:
    f.write('dummy model params')

# Mock the check_sha1 function to always return True
def check_sha1(fpath, sha1_hash):
    return True

# Mock the download function to do nothing
def download(url, path, overwrite=False):
    pass

# Inject mocks into the module namespace
import sys
current_module = sys.modules[__name__]
setattr(current_module, 'check_sha1', check_sha1)
setattr(current_module, 'download', download)

# Define ARGS for get_model_file
ARGS = {
    'name': 'arcface_r100_v1',
    'root': model_root
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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_model_file Ran
ran; our contract draft was wrong, not the code
code-trip/insightface-mxnet/python-package/insightface/model_zoo/model_store.py
code served (permissive licence) · get_code("16443b6e8ae03637")
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 str. 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 os
import tempfile
import shutil

# Mock the helper functions that would normally be imported
def check_sha1(file_path, sha1_hash):
    """Mock SHA1 check - always return True for testing"""
    return True

def download(url, path, overwrite=True):
    """Mock download function - create a dummy zip file"""
    import zipfile
    os.makedirs(os.path.dirname(path), exist_ok=True)
    # Create a dummy zip file with a .params file inside
    with zipfile.ZipFile(path, 'w') as zf:
        zf.writestr('arcface_r100_v1.params', 'dummy model params')

# Inject these into the module namespace for the function
import sys
current_module = sys.modules[__name__]
current_module.check_sha1 = check_sha1
current_module.download = download

# Create a temporary directory for testing
temp_dir = tempfile.mkdtemp()

ARGS = {
    'name': 'arcface_r100_v1',
    'root': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
im_proposal Ran
ran on a synthesized input
bleakie/RetinaDetection/rcnn/core/tester.py
code served (permissive licence) · get_code("239d9f658d3a3e68")
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
from unittest.mock import Mock

torch.manual_seed(0)

# Create a mock predictor that returns the expected outputs
class MockPredictor:
    def predict(self, data_batch):
        # Return a dict with 'rois_output' and 'rois_score'
        # rois_output has shape (batch_size, 5) where first column is batch index
        rois_output_array = np.array([
            [0.0, 10.0, 20.0, 30.0, 40.0],
            [1.0, 50.0, 60.0, 70.0, 80.0]
        ], dtype=np.float32)
        
        rois_score_array = np.array([
            [0.9, 0.1],
            [0.8, 0.2]
        ], dtype=np.float32)
        
        return {
            'rois_output': Mock(asnumpy=lambda: rois_output_array),
            'rois_score': Mock(asnumpy=lambda: rois_score_array)
        }

# Create a mock data_batch with data attribute
class MockDataBatch:
    def __init__(self, data):
        self.data = data

# Setup arguments
predictor = MockPredictor()

# data_batch should have a .data attribute that is iterable
# The function zips data_names with data_batch.data
data_batch = MockDataBatch([
    torch.randn(2, 4),
    torch.randn(2, 3)
])

data_names = ['features', 'labels']

scale = 2.0

ARGS = {
    'predictor': predictor,
    'data_batch': data_batch,
    'data_names': data_names,
    'scale': scale
}
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
image_to_torch Ran
ran on a synthesized input
hukkelas/DSFD-Pytorch-Inference/face_detection/torch_utils.py
code served (permissive licence) · get_code("8d94c589948db38f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 4, dtype family float, shape (1, 2, 4, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import torch

torch.manual_seed(0)

# Create a small uint8 image with shape (height, width, channels)
# Using distinct dimensions: 4 height, 3 width, 2 channels
image = np.random.randint(0, 256, size=(4, 3, 2), dtype=np.uint8)

device = 'cpu'

ARGS = {
    'image': image,
    'device': device
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (1, 2, 4, 3) — shape (1, 2, 4, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
index_fill Ran
ran on a synthesized input
GuoQuanhao/RetinaFace-Paddle/utils/box_utils.py
code served (permissive licence) · get_code("f5c78bcd24910e1d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (5, 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
import numpy as np

torch.manual_seed(0)

# Create a small tensor to be modified
input_tensor = torch.arange(10, dtype=torch.float32).reshape(5, 2)

# Create indices - list of positions to update
index = [0, 2, 4]

# Create update value - scalar that will be assigned
update = 99.0

ARGS = {
    "input": input_tensor,
    "index": index,
    "update": update
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (5, 2) — shape (5, 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
inference Ran
ran; our contract draft was wrong, not the code
yakhyo/retinaface-pytorch/detect.py
code served (permissive licence) · get_code("7c475f7ff32dc333")
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 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 torch.nn as nn

torch.manual_seed(0)

class SimpleDetectionModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv = nn.Conv2d(3, 16, kernel_size=3, padding=1)
        self.loc_head = nn.Linear(16 * 4 * 4, 8)
        self.conf_head = nn.Linear(16 * 4 * 4, 4)
        self.landmark_head = nn.Linear(16 * 4 * 4, 12)
    
    def forward(self, x):
        # x shape: (batch, 3, 4, 4)
        x = self.conv(x)  # (batch, 16, 4, 4)
        x = x.view(x.size(0), -1)  # (batch, 256)
        loc = self.loc_head(x).unsqueeze(0)  # (1, batch, 8)
        conf = self.conf_head(x).unsqueeze(0)  # (1, batch, 4)
        landmarks = self.landmark_head(x).unsqueeze(0)  # (1, batch, 12)
        return loc, conf, landmarks

model = SimpleDetectionModel()
image = torch.randn(2, 3, 4, 4)

ARGS = {
    "model": model,
    "image": image,
}
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 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
intersect Ran
ran; our contract draft was wrong, not the code
2275519762/RetinaFace_Pytorch/utils/box_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from NVlabs/AL-MDN, is pointer only (licence: NOASSERTION) · get_code("f01c5f73515770a5")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from NVlabs/AL-MDN, and GitHub could not classify the licence of that repository (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.

is_remote_url Ran
ran, violated the contract we drafted
StanislasBertrand/RetinaFace-tf2/src/retinafacetf2/utils/file_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from vis-nlp/ChartQA, is pointer only (licence: GPL-3.0) · get_code("899d1e2736236351")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from vis-nlp/ChartQA, and GPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. 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.

is_valid_landmark Ran
ran on a synthesized input
serengil/deepface/deepface/modules/detection.py
code served (permissive licence) · get_code("25c5c4be8e90a024")
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 bool, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from typing import Any, Dict, IO, List, Tuple, Union, Optional

# Set seed for reproducibility
torch.manual_seed(0)

# Construct valid arguments for is_valid_landmark
ARGS = {
    "coord": (50, 75),  # Valid (x, y) tuple within bounds
    "width": 100,
    "height": 150,
}
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 bool, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
l1loss_age Ran
ran on a synthesized input
sajjjadayobi/FaceLib/facelib/AgeGender/models/model.py
code served (permissive licence) · get_code("f17a7bd4407e0d00")
How it ran

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

import torch
import torch.nn.functional as F

torch.manual_seed(0)

# The function l1loss_age takes two parameters: input and targs
# It indexes both with [:, -1] (last column) and calls .mean() on the result
# So both need to be 2D tensors with at least 1 column

# Create input tensor: batch_size=2, features=4
input = torch.randn(2, 4)

# Create target tensor: batch_size=2, features=3 (different size to distinguish dimensions)
targs = torch.randn(2, 3)

ARGS = {
    "input": input,
    "targs": targs,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
l2_norm Ran
ran on a synthesized input
TinDang97/face_recognition/models/model_irse.py
code served (permissive licence) · get_code("4742e83323430543")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create a small tensor with shape (2, 4) - batch size 2, feature dim 4
x = torch.randn(2, 4)

# The function uses:
# - torch.norm(x, 2, axis, True) - computes L2 norm along specified axis
# - torch.div(x, norm) - divides x by the norm
# So x needs to be a tensor and axis needs to be an integer

ARGS = {
    "x": x,
    "axis": 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_info Ran
ran; our contract draft was wrong, not the code
peteryuX/retinaface-tf2/test_widerface.py
code served (permissive licence) · get_code("f0888a2af73aa016")
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 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 tempfile
import os

# Create a temporary directory and a temporary txt file with test data
temp_dir = tempfile.mkdtemp()
images_dir = os.path.join(temp_dir, 'images')
os.makedirs(images_dir, exist_ok=True)

txt_path = os.path.join(temp_dir, 'label.txt')

# Write test data to the txt file
with open(txt_path, 'w') as f:
    f.write("# image1.jpg\n")
    f.write("0.1 0.2 0.3\n")
    f.write("0.4 0.5 0.6\n")
    f.write("# image2.jpg\n")
    f.write("0.7 0.8 0.9\n")
    f.write("1.0 1.1 1.2\n")

ARGS = {
    'txt_path': txt_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — 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
load_weights Ran
ran on a synthesized input
StanislasBertrand/RetinaFace-tf2/src/retinafacetf2/networks/network_utils.py
code served (permissive licence) · get_code("162a2875f05e03d8")
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 numpy as np
import tempfile
import os

# Create a temporary directory for the test file
temp_dir = tempfile.mkdtemp()

# Create a sample weights dictionary with layer names and weight arrays
sample_weights = {
    'layer1': np.array([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]]),
    'layer2': np.array([7.0, 8.0, 9.0, 10.0]),
    'bias': np.array([0.1, 0.2, 0.3])
}

# Save the weights dictionary to an npy file
weight_file_path = os.path.join(temp_dir, 'test_weights.npy')
np.save(weight_file_path, sample_weights)

# Define the arguments for the function
ARGS = {
    'weight_file': weight_file_path
}
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 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
load_weights_to_network Ran
ran on a synthesized input
StanislasBertrand/RetinaFace-tf2/src/retinafacetf2/networks/network_utils.py
code served (permissive licence) · get_code("9d40efd483767314")
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 MockModel. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
try:
    import tensorflow as tf
    from tensorflow import keras
except ImportError:
    # Mock keras if not available
    class MockLayer:
        def __init__(self, name):
            self.name = name
            self._weights = []
        
        def set_weights(self, weights):
            self._weights = weights
    
    class MockModel:
        def __init__(self, layers):
            self.layers = layers
    
    keras = type('keras', (), {'layers': type('layers', (), {})()})()

# Create a mock Keras model with layers
class MockLayer:
    def __init__(self, name):
        self.name = name
        self._weights = []
    
    def set_weights(self, weights):
        self._weights = weights

class MockModel:
    def __init__(self, layers):
        self.layers = layers

# Create layers: one batch norm, one dense layer with bias, one without bias
bn_layer = MockLayer('bn_layer_1')
dense_layer_with_bias = MockLayer('dense_layer_1')
dense_layer_no_bias = MockLayer('dense_layer_2')

model = MockModel([bn_layer, dense_layer_with_bias, dense_layer_no_bias])

# Create weights_dict with proper structure
weights_dict = {
    'bn_layer_1': {
        'mean': np.random.randn(4).astype(np.float32),
        'var': np.random.randn(4).astype(np.float32),
        'scale': np.ones(4, dtype=np.float32),
        'bias': np.zeros(4, dtype=np.float32)
    },
    'dense_layer_1': {
        'weights': np.random.randn(5, 3).astype(np.float32),
        'bias': np.random.randn(3).astype(np.float32)
    },
    'dense_layer_2': {
        'weights': np.random.randn(6, 4).astype(np.float32)
    }
}

ARGS = {
    'model': model,
    'weights_dict': weights_dict
}
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 MockModel — a MockModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a MockModel 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
logger_init Ran
ran on a synthesized input
iChenning/facedetection/utils/log.py
code served (permissive licence) · get_code("1f87f556e0e70e55")
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 Logger. 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 os
import tempfile
import logging

# Create a temporary directory for the logger to write to
temp_dir = tempfile.mkdtemp()

# Define ARGS with the required parameter
ARGS = {
    'folder_dir': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Logger — a Logger output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Logger 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
nms Ran
ran on a synthesized input
hhj1897/face_detection/ibug/face_detection/s3fd/utils.py
code served (permissive licence) · get_code("d862913455d35d7f")
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

torch.manual_seed(0)

# Create bounding boxes in format [x1, y1, x2, y2]
# Shape: [num_priors, 4]
boxes = torch.tensor([
    [10.0, 20.0, 30.0, 40.0],
    [15.0, 25.0, 35.0, 45.0],
], dtype=torch.float32)

# Create scores for each box
# Shape: [num_priors]
scores = torch.tensor([0.9, 0.8], dtype=torch.float32)

# Define arguments dict
ARGS = {
    'boxes': boxes,
    'scores': scores,
    'overlap': 0.5,
    'top_k': 200,
}
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 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
nms_np Ran
ran on a synthesized input
hhj1897/face_detection/ibug/face_detection/s3fd/utils.py
code served (permissive licence) · get_code("93db22d0260a0326")
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

torch.manual_seed(0)

# Create small bounding boxes: [num_priors, 4] format (x1, y1, x2, y2)
boxes = torch.tensor([
    [10.0, 20.0, 50.0, 60.0],
    [15.0, 25.0, 55.0, 65.0],
    [100.0, 100.0, 150.0, 150.0],
], dtype=torch.float32)

# Create scores for each box: [num_priors]
scores = torch.tensor([0.9, 0.8, 0.7], dtype=torch.float32)

# Define parameters
overlap = 0.5
top_k = 200

ARGS = {
    "boxes": boxes,
    "scores": scores,
    "overlap": overlap,
    "top_k": top_k,
}
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
normalize_input Ran
ran on a synthesized input
serengil/deepface/deepface/modules/preprocessing.py
code served (permissive licence) · get_code("207ca801fd14f759")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 3, dtype family float, shape (4, 3, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create a small test image as a numpy array
# Shape: (4, 3, 3) - batch of 4 images, 3x3 pixels, 3 channels (RGB)
img = np.random.rand(4, 3, 3).astype(np.float32)

ARGS = {
    'img': img,
    'normalization': 'base'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family float, shape (4, 3, 3) — shape (4, 3, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
pa_multibox Ran
ran on a synthesized input
hukkelas/DSFD-Pytorch-Inference/face_detection/dsfd/face_ssd.py
code served (permissive licence) · get_code("e4d1c0f6da7d33f8")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 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

torch.manual_seed(0)

# Analyze the function:
# - output_channels: iterated with enumerate, so needs to be a sequence of integers
# - mbox_cfg: indexed with k (integer indices), so needs to be a sequence of integers
# - num_classes: not used in the function body, but required parameter

ARGS = {
    "output_channels": [512, 1024, 2048],
    "mbox_cfg": [4, 6, 6],
    "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 tuple, ndim 2, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
point_form Ran
ran; our contract draft was wrong, not the code
2275519762/RetinaFace_Pytorch/utils/box_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from NVlabs/AL-MDN, is pointer only (licence: NOASSERTION) · get_code("3bea1addc1e14162")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from NVlabs/AL-MDN, and GitHub could not classify the licence of that repository (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.

py_cpu_nms Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
2275519762/RetinaFace_Pytorch/utils/nms/py_cpu_nms.py
pointer only (licence: MIT); the stored copy get_code serves, from Anymake/DRN_CVPR2020, is pointer only (licence: GPL-2.0) · get_code("5502a789c2da0e5e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from Anymake/DRN_CVPR2020, and GPL-2.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. 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.

py_cpu_nms Ran
ran on a synthesized input
hhj1897/face_detection/ibug/face_detection/retina_face/py_cpu_nms.py
code served (permissive licence) · get_code("772670c1695ac7b9")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Construct a small detections array with format [x1, y1, x2, y2, score]
# Creating 5 detections with distinct coordinates and scores
dets = np.array([
    [10.0, 20.0, 50.0, 60.0, 0.9],
    [15.0, 25.0, 55.0, 65.0, 0.8],
    [100.0, 120.0, 150.0, 160.0, 0.85],
    [12.0, 22.0, 52.0, 62.0, 0.7],
    [200.0, 220.0, 250.0, 260.0, 0.95],
], dtype=np.float32)

# NMS threshold - detections with IoU above this will be suppressed
thresh = 0.3

# Keep only top_k detections (after NMS filtering)
top_k = 3

ARGS = {
    "dets": dets,
    "thresh": thresh,
    "top_k": top_k,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family int — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
read_label Ran
ran; our contract draft was wrong, not the code
yangyucheng000/RetinaFace/src/retinaface_eval.py
code served (permissive licence) · get_code("bf91d15cb1c6a6d8")
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 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 tempfile
import os

# Create a temporary label file with test data
temp_dir = tempfile.mkdtemp()
label_path = os.path.join(temp_dir, "label.txt")

# Write test label file with image paths
with open(label_path, 'w') as f:
    f.write("# /path/to/image1.jpg\n")
    f.write("# /path/to/image2.png\n")
    f.write("# /path/to/image3.jpg\n")
    f.write("some_other_line\n")
    f.write("# /path/to/image4.jpg\n")

ARGS = {
    "label_path": label_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — 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
read_pred_file Ran
ran; our contract draft was wrong, not the code
2275519762/RetinaFace_Pytorch/widerface_evaluate/evaluation.py
code served (permissive licence) · get_code("5e211291c72bb0bd")
How it ran

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

import numpy as np
import tempfile
import os

# Create a temporary prediction file
temp_dir = tempfile.mkdtemp()
pred_filepath = os.path.join(temp_dir, "pred.txt")

# Write a sample prediction file with the expected format
with open(pred_filepath, 'w') as f:
    f.write("path/to/image.jpg\n")
    f.write("\n")
    f.write("0.1 0.2 0.3 0.4 0.9\n")
    f.write("0.5 0.6 0.7 0.8 0.95\n")
    f.write("0.15 0.25 0.35 0.45 0.85\n")

ARGS = {
    'filepath': pred_filepath
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Showing the first 60. An agent gets every one from get_harvested_code_for_paper("1905.00641").

Also present in other papers' code

Of 91 samples lifted from this paper's linked repositories, 91 carry a function identity and 0 do not. Of the 90 distinct functions with an identity, 39 are also present in other papers whose code we hold (15 with at least one paper dated earlier, 8 the same month, 39 later) and 51 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
conv3x3 686 other papers
119 dated earlier, 12 the same month, 555 later · 31 only as an assembly brick
1406.2661 (2014-06)
Generative Adversarial Networks
1409.0575 (2014-09)
ImageNet Large Scale Visual Recognition Challenge
1412.6572 (2014-12)
Explaining and Harnessing Adversarial Examples
and 683 more not named here
conv1x1 431 other papers
33 dated earlier, 2 the same month, 395 later, 1 undated · 38 only as an assembly brick
1409.1556 (2014-09)
Very Deep Convolutional Networks for Large-Scale Image Recognition
1506.01186 (2015-06)
Cyclical Learning Rates for Training Neural Networks
1507.05717 (2015-07)
An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition
and 428 more not named here
conv3x3 312 other papers
14 dated earlier, 297 later, 1 undated · 31 only as an assembly brick
1409.1556 (2014-09)
Very Deep Convolutional Networks for Large-Scale Image Recognition
1507.05717 (2015-07)
An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition
and 309 more not named here
resnet18 89 other papers
15 dated earlier, 2 the same month, 72 later
1705.09759 (2017-05)
CASENet: Deep Category-Aware Semantic Edge Detection
1801.07698 (2018-01)
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
1803.00839 (2018-03)
Pose-Robust Face Recognition via Deep Residual Equivariant Mapping
and 86 more not named here
resnet34 53 other papers
14 dated earlier, 1 the same month, 38 later
1705.09759 (2017-05)
CASENet: Deep Category-Aware Semantic Edge Detection
1801.07698 (2018-01)
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
1803.00839 (2018-03)
Pose-Robust Face Recognition via Deep Residual Equivariant Mapping
and 50 more not named here
str2bool 18 other papers
18 later
1908.01491 (2019-08)
Pixel2Mesh++: Multi-View 3D Mesh Generation via Deformation
1909.13589 (2019-09)
Domain Adaptation for Semantic Segmentation with Maximum Squares Loss
2002.12416 (2020-02)
Learning in the Frequency Domain
and 15 more not named here
intersect 14 other papers
6 dated earlier, 8 later
1409.1556 (2014-09)
Very Deep Convolutional Networks for Large-Scale Image Recognition
1512.02325 (2015-12)
SSD: Single Shot MultiBox Detector
and 11 more not named here
is_remote_url 14 other papers
14 later
1910.13461 (2019-10)
BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension
2004.10964 (2020-04)
Don't Stop Pretraining: Adapt Language Models to Domains and Tasks
2104.08066 (2021-04)
Effect of Visual Extensions on Natural Language Understanding in Vision-and-Language Models
and 11 more not named here
point_form 12 other papers
5 dated earlier, 7 later
1409.1556 (2014-09)
Very Deep Convolutional Networks for Large-Scale Image Recognition
1512.02325 (2015-12)
SSD: Single Shot MultiBox Detector
and 9 more not named here
short_hash 11 other papers
2 dated earlier, 1 the same month, 8 later
1803.08904 (2018-03)
Context Encoding for Semantic Segmentation
1809.02983 (2018-09)
Dual Attention Network for Scene Segmentation
1905.06368 (2019-05)
Collaborative Global-Local Networks for Memory-Efficient Segmentation of Ultra-High Resolution Images
and 8 more not named here
center_size 9 other papers
4 dated earlier, 5 later
1409.1556 (2014-09)
Very Deep Convolutional Networks for Large-Scale Image Recognition
1512.02325 (2015-12)
SSD: Single Shot MultiBox Detector
and 6 more not named here
py_cpu_nms 9 other papers
2 dated earlier, 1 the same month, 6 later
1506.01497 (2015-06)
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
1812.00155 (2018-12)
Learning RoI Transformer for Detecting Oriented Objects in Aerial Images
1905.02244 (2019-05)
Searching for MobileNetV3
and 6 more not named here
densenet121 8 other papers
8 later
1912.00385 (2019-12)
The Group Loss for Deep Metric Learning
2010.11742 (2020-10)
Learning Black-Box Attackers with Transferable Priors and Query Feedback
2104.01231 (2021-04)
Diverse Gaussian Noise Consistency Regularization for Robustness and Uncertainty Calibration
and 5 more not named here
densenet161 8 other papers
1 the same month, 7 later
1905.11926 (2019-05)
Network Deconvolution
1912.00385 (2019-12)
The Group Loss for Deep Metric Learning
2010.11742 (2020-10)
Learning Black-Box Attackers with Transferable Priors and Query Feedback
and 5 more not named here
densenet169 8 other papers
1 the same month, 7 later
1905.11926 (2019-05)
Network Deconvolution
1912.00385 (2019-12)
The Group Loss for Deep Metric Learning
2010.11742 (2020-10)
Learning Black-Box Attackers with Transferable Priors and Query Feedback
and 5 more not named here
relu6 6 other papers
5 dated earlier, 1 later
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition
1612.03716 (2016-12)
COCO-Stuff: Thing and Stuff Classes in Context
1704.04861 (2017-04)
MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
and 3 more not named here
conv_bn 4 other papers
1 dated earlier, 3 later
1703.07332 (2017-03)
How far are we from solving the 2D &amp; 3D Face Alignment problem? (and a dataset of 230,000 3D facial landmarks)
2203.11593 (2022-03)
Unified Negative Pair Generation toward Well-discriminative Feature Space for Face Recognition
2303.17176 (2023-03)
A View From Somewhere: Human-Centric Face Representations
and 1 more not named here
conv_bn_no_relu 4 other papers
1 dated earlier, 3 later
1703.07332 (2017-03)
How far are we from solving the 2D &amp; 3D Face Alignment problem? (and a dataset of 230,000 3D facial landmarks)
2203.11593 (2022-03)
Unified Negative Pair Generation toward Well-discriminative Feature Space for Face Recognition
2303.17176 (2023-03)
A View From Somewhere: Human-Centric Face Representations
and 1 more not named here
IOU 3 other papers
2 dated earlier, 1 later
1801.07698 (2018-01)
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
2003.06814 (2020-03)
Towards Face Encryption by Generating Adversarial Identity Masks
check_keys 3 other papers
3 later
2006.14749 (2020-06)
Deepfake Detection using Spatiotemporal Convolutional Networks
2303.17176 (2023-03)
A View From Somewhere: Human-Centric Face Representations
2411.02537 (2024-11)
INQUIRE: A Natural World Text-to-Image Retrieval Benchmark

Shown: the 20 functions present in the most other papers. 19 more shared functions are not listed here.

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

Repositories linked to this paper

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

Though tremendous strides have been made in uncontrolled face detection, accurate and efficient face localisation in the wild remains an open challenge. This paper presents a robust single-stage face detector, named RetinaFace, which performs pixel-wise face localisation on various scales of faces by taking advantages of joint extra-supervised and self-supervised multi-task learning. Specifically, We make contributions in the following five aspects: (1) We manually annotate five facial landmarks on the WIDER FACE dataset and observe significant improvement in hard face detection with the assistance of this extra supervision signal. (2) We further add a self-supervised mesh decoder branch for predicting a pixel-wise 3D shape face information in parallel with the existing supervised branches. (3) On the WIDER FACE hard test set, RetinaFace outperforms the state of the art average precision (AP) by 1.1% (achieving AP equal to 91.4%). (4) On the IJB-C test set, RetinaFace enables state of the art methods (ArcFace) to improve their results in face verification (TAR=89.59% for FAR=1e-6). (5) By employing light-weight backbone networks, RetinaFace can run real-time on a single CPU core for a VGA-resolution image. Extra annotations and code have been made available at: https://github.com/deepinsight/insightface/tree/master/RetinaFace.

For agents

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

get_harvested_code_for_paper("1905.00641")
get_code_for_paper("1905.00641")
have("1905.00641")

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

curl https://syntology.ai/api/ran/1905.00641.json

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

[![Syntology run record](https://syntology.ai/api/ran/1905.00641.svg)](https://syntology.ai/paper/1905.00641)

Connect an agent — have() is free.