SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1906.07155 · 2019

MMDetection: Open MMLab Detection Toolbox and Benchmark

arXiv · PDF · Open in the Atlas

Code that ran

62 of 82 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; 4 honoured the contract we drafted, 3 violated it and 30 have no contract verdict; 25 failures of our instrument; 20 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/1906.07155.json · badge: /api/ran/1906.07155.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
HAOCHENYE/yehc_mmdet community (classifier label)
classifier basis not recorded on the link
6 ran · 3 no recorded run
sathishkumark27/mmdetection-icevision community (classifier label)
classifier basis not recorded on the link
6 ran · 0 no recorded run
YeLyuUT/MMDET community (classifier label)
classifier basis not recorded on the link
5 ran · 5 no recorded run
FayCai/framework community (classifier label)
classifier basis not recorded on the link
5 ran · 5 no recorded run
nithinksath96/MMdetection_TensorRT_FP16 community (classifier label)
classifier basis not recorded on the link
5 ran · 0 no recorded run
abcxs/polyrnn community (classifier label)
classifier basis not recorded on the link
5 ran · 0 no recorded run
dereyly/mmdet_sota community (classifier label)
classifier basis not recorded on the link
4 ran · 3 no recorded run
fehmikahraman/mmdetection community (classifier label)
classifier basis not recorded on the link
4 ran · 2 no recorded run
floatingstarZ/loc_cls_exp community (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
mengfu188/mmdetection.bak community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
RuiminChen/centernet community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
13952522076/mmdet0 community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
coldsummerday/mmdetection-zhou community (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
jichengYUAN/mmdetectionCust community (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
silist/mmdet_dev community (classifier label)
classifier basis not recorded on the link
1 ran · 2 no recorded run
lhwcv/mmdet extension (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
hSDB-instrument/hsdb-mmdetection community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
ZZR8066/mmdetection-GaofenCompetition community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
JHuang-CV/mmdetectionV2 community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
FunctionStatusWhere it lives
ScaleRatioToStageOutChannels Ran
ran on a synthesized input
coldsummerday/mmdetection-zhou/zhoutools/shufflenetv2.py
code served (permissive licence) · get_code("c7b1ef4be853562d")
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 torch

def ScaleRatioToStageOutChannels(ratio):
    if ratio == 0.5:
        return [24, 48, 96, 192]
    elif ratio == 1:
        return [24, 116, 232, 464]
    elif ratio == 1.5:
        return [24, 176, 352, 704]
    elif ratio == 2.0:
        return [24, 244, 488, 976]

ARGS = {
    "ratio": 1.0
}
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 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
anchor_inside_flags Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
HAOCHENYE/yehc_mmdet/mmdet/core/anchor/utils.py
code served (permissive licence) · get_code("cc1d87b0a0678212")
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 1, dtype family float, shape (2,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create flat_anchors: shape (n, 4) representing anchor boxes [x1, y1, x2, y2]
flat_anchors = torch.tensor([
    [10.0, 20.0, 50.0, 80.0],
    [5.0, 15.0, 100.0, 120.0]
], dtype=torch.float32)

# Create valid_flags: boolean tensor of same length as flat_anchors
valid_flags = torch.tensor([True, True], dtype=torch.bool)

# Image shape (height, width, channels) - only first 2 elements used
img_shape = (256, 512, 3)

# allowed_border parameter
allowed_border = 0

ARGS = {
    'flat_anchors': flat_anchors,
    'valid_flags': valid_flags,
    'img_shape': img_shape,
    'allowed_border': allowed_border
}
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 1, dtype family float, shape (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
anchor_inside_flags Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
dereyly/mmdet_sota/mmdet/core/anchor/utils.py
code served (permissive licence) · get_code("cb9ac602e0a8f08a")
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 1, dtype family float, shape (2,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# flat_anchors: tensor of shape (N, 4) with values [x1, y1, x2, y2]
# Each row represents an anchor box with coordinates
flat_anchors = torch.tensor([
    [10.0, 20.0, 50.0, 60.0],
    [5.0, 15.0, 45.0, 55.0],
], dtype=torch.float32)

# valid_flags: boolean tensor of shape (N,) indicating which anchors are valid
valid_flags = torch.tensor([True, True], dtype=torch.bool)

# img_shape: tuple or list of image dimensions (height, width, channels)
# The function uses img_shape[:2] to get height and width
img_shape = (100, 120, 3)

# allowed_border: integer controlling border tolerance
allowed_border = 0

ARGS = {
    "flat_anchors": flat_anchors,
    "valid_flags": valid_flags,
    "img_shape": img_shape,
    "allowed_border": allowed_border,
}
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 1, dtype family float, shape (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
auto_fp16 Ran
ran; our contract draft was wrong, not the code
nithinksath96/MMdetection_TensorRT_FP16/mmdet/core/fp16/decorators.py
code served (permissive licence) · get_code("4020896b8cf7566a")
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 torch
import torch.nn as nn
import functools
from inspect import getfullargspec

torch.manual_seed(0)

def cast_tensor_type(input_, src_type, dst_type):
    """Helper function to cast tensor types."""
    if isinstance(input_, torch.Tensor):
        if input_.dtype == src_type:
            return input_.to(dst_type)
    return input_

class SimpleModule(nn.Module):
    """A minimal nn.Module for testing the decorator."""
    def __init__(self):
        super().__init__()
        self.fp16_enabled = True
        self.linear = nn.Linear(4, 3)
    
    @staticmethod
    def decorated_method_all(x, y):
        """Method decorated with auto_fp16 applying to all arguments."""
        return x + y
    
    @staticmethod
    def decorated_method_selective(pred, others):
        """Method decorated with auto_fp16 applying only to pred."""
        return pred + others

test_module = SimpleModule()

x_tensor = torch.randn(2, 4, dtype=torch.float32)
y_tensor = torch.randn(2, 4, dtype=torch.float32)
pred_tensor = torch.randn(2, 3, dtype=torch.float32)
others_tensor = torch.randn(2, 3, dtype=torch.float32)

ARGS = {
    'apply_to': ('pred',),
    'out_fp32': 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 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
backbone_head_filter Ran
ran on a synthesized input
floatingstarZ/loc_cls_exp/CONTROL_CODE/config.py
code served (permissive licence) · get_code("c943ebd63eda5d0a")
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 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)

class SimpleBackbone(nn.Module):
    def __init__(self):
        super().__init__()
        self.backbone = nn.Linear(4, 3)
        self.neck = nn.Linear(3, 2)
        self.head = nn.Linear(2, 1)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.backbone_layer = nn.Linear(4, 3)
        self.neck_layer = nn.Linear(3, 2)
        self.head_layer = nn.Linear(2, 1)
        self.other_layer = nn.Linear(1, 1)
    
    def forward(self, x):
        x = self.backbone_layer(x)
        x = self.neck_layer(x)
        x = self.head_layer(x)
        return x

model = SimpleModel()

ARGS = {
    "model": model
}
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 1, dtype family float — 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
bbox_flip Ran
ran on a synthesized input
HAOCHENYE/yehc_mmdet/mmdet/core/bbox/transforms.py
code served (permissive licence) · get_code("6a29f5a7da19f7ae")
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, 8) 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 bboxes tensor with shape (..., 4*k)
# Using shape (2, 8) means batch of 2, with 2 bboxes (4*2=8 coordinates)
bboxes = torch.tensor([
    [10.0, 20.0, 30.0, 40.0, 50.0, 60.0, 70.0, 80.0],
    [15.0, 25.0, 35.0, 45.0, 55.0, 65.0, 75.0, 85.0]
])

# Image shape as (height, width)
img_shape = (100, 120)

# Direction for flipping
direction = 'horizontal'

ARGS = {
    'bboxes': bboxes,
    'img_shape': img_shape,
    'direction': direction
}
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, 8) — shape (2, 8) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
bbox_flip Ran
ran on a synthesized input
dereyly/mmdet_sota/mmdet/core/bbox/transforms.py
code served (permissive licence) · get_code("1d95f6fc3e21b377")
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, 8) 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 bboxes tensor with shape (..., 4*k)
# Using shape (2, 8) where 8 = 4*2 (2 bounding boxes, each with 4 coordinates)
# Format: (x1, y1, x2, y2, x1, y1, x2, y2) for 2 boxes
bboxes = torch.tensor([
    [10.0, 20.0, 50.0, 60.0, 15.0, 25.0, 55.0, 65.0],
    [30.0, 40.0, 70.0, 80.0, 35.0, 45.0, 75.0, 85.0]
])

# Image shape as (height, width)
img_shape = (100, 120)

# Direction parameter
direction = 'horizontal'

ARGS = {
    'bboxes': bboxes,
    'img_shape': img_shape,
    'direction': direction
}
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, 8) — shape (2, 8) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
bbox_overlaps Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
YeLyuUT/MMDET/mmdet/core/bbox/geometry.py
code served (permissive licence) · get_code("d65ea05c493bc2d8")
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, 2) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create two sets of bounding boxes
# Format: (x1, y1, x2, y2) where x1, y1 is top-left and x2, y2 is bottom-right
bboxes1 = torch.tensor([
    [10.0, 10.0, 50.0, 50.0],
    [20.0, 20.0, 60.0, 60.0]
], dtype=torch.float32)

bboxes2 = torch.tensor([
    [15.0, 15.0, 55.0, 55.0],
    [30.0, 30.0, 70.0, 70.0]
], dtype=torch.float32)

ARGS = {
    'bboxes1': bboxes1,
    'bboxes2': bboxes2,
    'mode': 'iou',
    'is_aligned': False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 2) — shape (2, 2) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
bbox_overlaps Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
HAOCHENYE/yehc_mmdet/mmdet/core/evaluation/bbox_overlaps.py
code served (permissive licence) · get_code("c21eb325c33ec5d7")
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, 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 small bounding boxes for testing
# Format: [x1, y1, x2, y2]
bboxes1 = np.array([
    [10, 10, 50, 50],
    [20, 20, 60, 60]
], dtype=np.int64)

bboxes2 = np.array([
    [30, 30, 70, 70],
    [40, 40, 80, 80],
    [5, 5, 25, 25]
], dtype=np.int64)

ARGS = {
    'bboxes1': bboxes1,
    'bboxes2': bboxes2,
    'mode': 'iou',
    'eps': 1e-06
}
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, 3) — shape (2, 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
cast_tensor_type Ran
ran; our contract draft was wrong, not the code
nithinksath96/MMdetection_TensorRT_FP16/mmdet/core/fp16/utils.py
code served (permissive licence) · get_code("b5bfd1f48ed5c4d9")
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 abc

torch.manual_seed(0)

# Create test data structures with nested tensors and arrays
test_tensor = torch.randn(2, 4)
test_numpy = np.array([[1.0, 2.0], [3.0, 4.0]])
test_string = "test_string"

# Create a nested mapping (dict)
test_dict = {
    "tensor": torch.randn(2, 4),
    "nested": {
        "array": np.array([1, 2, 3]),
        "string": "nested_string"
    }
}

# Create a nested iterable (tuple with mixed types)
test_tuple = (
    torch.randn(2, 4),
    np.array([5, 6, 7]),
    "tuple_string"
)

# Create a list with mixed nested content
test_list = [
    torch.randn(2, 4),
    {"inner_tensor": torch.randn(3, 2)},
    "list_item"
]

ARGS = {
    "inputs": test_list,
    "src_type": torch.float32,
    "dst_type": torch.float64
}
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
change_light Ran
ran on a synthesized input
sathishkumark27/mmdetection-icevision/augmentations/augmentations.py
code served (permissive licence) · get_code("a7c92643810674da")
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 int, 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 cv2
import numpy as np

# Create a small RGB image (height=4, width=3, channels=3)
# Using uint8 values in valid RGB range [0, 255]
image = np.random.randint(0, 256, size=(4, 3, 3), dtype=np.uint8)

# Create a coefficient to brighten the image
coeff = 1.2

ARGS = {
    "image": image,
    "coeff": coeff,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, 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 int, 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
channel_shuffle Ran
ran; our contract draft was wrong, not the code
coldsummerday/mmdetection-zhou/zhoutools/shufflenetv2.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ddghost/new_darts, is pointer only (licence: NONE) · get_code("b9da06d4f527dd6c")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from ddghost/new_darts, 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.

check_anno_index Ran
ran on a synthesized input
jichengYUAN/mmdetectionCust/utils.py
code served (permissive licence) · get_code("997c5298e5d40679")
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 1, 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 json
import tempfile
import os

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

# Create a COCO format annotation dictionary with categories starting at id 0
anno_dict = {
    'images': [
        {'id': 1, 'file_name': 'image1.jpg'},
        {'id': 2, 'file_name': 'image2.jpg'}
    ],
    'annotations': [
        {'id': 1, 'image_id': 1, 'category_id': 0, 'area': 100},
        {'id': 2, 'image_id': 2, 'category_id': 1, 'area': 200}
    ],
    'categories': [
        {'id': 0, 'name': 'cat'},
        {'id': 1, 'name': 'dog'}
    ]
}

# Write the annotation file to disk
anno_path = os.path.join(temp_dir, 'annotations.json')
with open(anno_path, 'w') as f:
    json.dump(anno_dict, f)

ARGS = {
    'path_to_anno': anno_path
}
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 tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
check_link Ran
ran; our contract draft was wrong, not the code
fehmikahraman/mmdetection/.dev_scripts/check_links.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from opendatalab/MLS-BRN, is pointer only (licence: NONE) · get_code("4e2dcdfe578946a0")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from opendatalab/MLS-BRN, 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.

check_url Ran
ran; our contract draft was wrong, not the code
fehmikahraman/mmdetection/.dev_scripts/check_links.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from opendatalab/MLS-BRN, is pointer only (licence: NONE) · get_code("cc93fdc812eca8f7")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from opendatalab/MLS-BRN, 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.

convert Ran
ran, violated the contract we drafted
lhwcv/mmdet/CZUR/head_data_preprocess/convert_to_yolo_txt.py
code served (permissive licence) · get_code("ede22d5f3c804e5b")
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 tuple, ndim 1, 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 numpy as np

torch.manual_seed(0)

# The function expects:
# - size: a tuple/list with 2 elements (width, height)
# - box: a tuple/list with 4 elements representing bounding box coordinates
#   Based on the code: box[0], box[1] are x-coordinates, box[2], box[3] are y-coordinates
#   It appears to convert from absolute coordinates to normalized YOLO format

size = (640, 480)  # width, height - distinct dimensions
box = (100, 200, 150, 300)  # xmin, xmax, ymin, ymax

ARGS = {
    'size': size,
    'box': box
}
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 tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
convert_bbox_to_z Ran
ran; our contract draft was wrong, not the code
mengfu188/mmdetection.bak/sort.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from chetanchawla/Sort-Tracker, is pointer only (licence: NONE) · get_code("90a45e51733f1779")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from chetanchawla/Sort-Tracker, 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.

convert_x_to_bbox Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
mengfu188/mmdetection.bak/sort.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from chetanchawla/Sort-Tracker, is pointer only (licence: NONE) · get_code("45940bb4051db020")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from chetanchawla/Sort-Tracker, 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.

create_image_id_infos Ran
ran on a synthesized input
FayCai/framework/convert_and_ensemble.py
code served (permissive licence) · get_code("22c3b745286482ec")
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 copy

ARGS = {
    'js_dict': {
        'images': [
            {
                'image_id': 'img_001',
                'file_name': 'image1.jpg',
                'width': 640,
                'height': 480
            },
            {
                'image_id': 'img_002',
                'file_name': 'image2.jpg',
                'width': 800,
                'height': 600
            }
        ]
    },
    'using_str': True
}
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 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
create_image_id_submit_infos Ran
ran on a synthesized input
FayCai/framework/convert_and_ensemble.py
code served (permissive licence) · get_code("39e468e6fe7823f1")
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 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 json
import tempfile
import os

# Create temporary JSON files to simulate model submit files
temp_dir = tempfile.mkdtemp()

# Create first submit file
submit_file_1 = os.path.join(temp_dir, "submit_1.json")
submit_data_1 = [
    {"image_id": 0, "score": 0.9, "label": "cat"},
    {"image_id": 1, "score": 0.8, "label": "dog"},
    {"image_id": 0, "score": 0.7, "label": "bird"}
]
with open(submit_file_1, 'w', encoding='utf-8') as f:
    json.dump(submit_data_1, f)

# Create second submit file
submit_file_2 = os.path.join(temp_dir, "submit_2.json")
submit_data_2 = [
    {"image_id": 1, "score": 0.85, "label": "cat"},
    {"image_id": 0, "score": 0.75, "label": "dog"}
]
with open(submit_file_2, 'w', encoding='utf-8') as f:
    json.dump(submit_data_2, f)

# Create id_to_str mapping
id_to_str = {
    0: "image_001",
    1: "image_002"
}

# Define arguments
ARGS = {
    "model_submit_files": [submit_file_1, submit_file_2],
    "id_to_str": id_to_str
}
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 list, ndim 1, dtype family float — 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
create_relation Ran
ran on a synthesized input
FayCai/framework/convert_and_ensemble.py
code served (permissive licence) · get_code("5355ba1134246e5c")
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 copy

# Construct a valid js_dict argument for create_relation
# The function expects a dict with an 'images' key containing a list of image info dicts
# Each image info dict should have 'image_id' and 'id' keys

js_dict = {
    'images': [
        {
            'image_id': 'img_001',
            'id': 1,
            'other_field': 'some_value'
        },
        {
            'image_id': 'img_002',
            'id': 2,
            'other_field': 'another_value'
        },
        {
            'image_id': 'img_003',
            'id': 3,
            'other_field': 'third_value'
        }
    ]
}

ARGS = {
    'js_dict': js_dict
}
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 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
decoder_filter Ran
ran on a synthesized input
floatingstarZ/loc_cls_exp/CONTROL_CODE/config.py
code served (permissive licence) · get_code("f84e45e38bf1bf32")
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 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)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.encoder = nn.Linear(4, 3)
        self.decoder = nn.Linear(3, 2)
        self.decoder_layer2 = nn.Linear(2, 1)
        self.other_param = nn.Linear(1, 1)
    
    def forward(self, x):
        x = self.encoder(x)
        x = self.decoder(x)
        x = self.decoder_layer2(x)
        x = self.other_param(x)
        return x

model = SimpleModel()

ARGS = {
    "model": model
}
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 1, dtype family float — 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
ensure_rng Ran
ran; our contract draft was wrong, not the code
HAOCHENYE/yehc_mmdet/mmdet/core/bbox/demodata.py
code served (permissive licence) · get_code("272ba15451a11158")
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 RandomState. 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

# Test case 1: None argument (uses global rng)
args_none = None

# Test case 2: Integer seed
args_int = 42

# Test case 3: RandomState object
args_randomstate = np.random.RandomState(123)

# Create ARGS dict with all parameter combinations
ARGS = {
    'rng': None
}
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 RandomState — a RandomState output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a RandomState 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
ensure_rng Ran
ran; our contract draft was wrong, not the code
dereyly/mmdet_sota/mmdet/core/bbox/demodata.py
code served (permissive licence) · get_code("8f43d2c26227a141")
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 RandomState. 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

# Test case 1: None (uses global rng)
case_1_rng = None

# Test case 2: integer seed
case_2_rng = 42

# Test case 3: RandomState object
case_3_rng = np.random.RandomState(123)

ARGS = {
    "rng": case_1_rng,
}
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 RandomState — a RandomState output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a RandomState 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
expand_target Ran
ran on a synthesized input
YeLyuUT/MMDET/mmdet/core/bbox/bbox_target.py
code served (permissive licence) · get_code("a57ba57f6dde621b")
How it ran

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

import torch

torch.manual_seed(0)

# Create small tensors with batch_size=2, bbox_dim=4
batch_size = 2
num_classes = 3

# bbox_targets: (batch_size, 4) - bounding box regression targets
bbox_targets = torch.randn(batch_size, 4)

# bbox_weights: (batch_size, 4) - weights for each bbox coordinate
bbox_weights = torch.ones(batch_size, 4)

# labels: (batch_size,) - class labels, values should be > 0 for some samples
# Using values in range [0, num_classes) where > 0 means object is present
labels = torch.tensor([0, 1], dtype=torch.long)

ARGS = {
    "bbox_targets": bbox_targets,
    "bbox_weights": bbox_weights,
    "labels": labels,
    "num_classes": num_classes,
}
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 3, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
filter_for_annotations Ran
ran on a synthesized input
sathishkumark27/mmdetection-icevision/albu/augmenter.py
code served (permissive licence) · get_code("7ec44c9d38cd5d14")
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 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 os
import tempfile
import shutil

# Create a temporary directory structure for testing
temp_root = tempfile.mkdtemp()
annotation_subdir = "annotations"
annotation_path = os.path.join(temp_root, annotation_subdir)
os.makedirs(annotation_path, exist_ok=True)

# Create some test .tsv files
test_files = ["data1.tsv", "data2.tsv", "results.tsv"]
for filename in test_files:
    filepath = os.path.join(annotation_path, filename)
    with open(filepath, 'w') as f:
        f.write("col1\tcol2\tcol3\n")
        f.write("val1\tval2\tval3\n")

# Create a non-tsv file that should be filtered out
non_tsv_file = os.path.join(annotation_path, "readme.txt")
with open(non_tsv_file, 'w') as f:
    f.write("This is not a TSV file")

ARGS = {
    "root": temp_root,
    "annotation_dir": annotation_subdir,
}
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 list, ndim 1, dtype family float — 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
force_fp32 Ran
ran; our contract draft was wrong, not the code
nithinksath96/MMdetection_TensorRT_FP16/mmdet/core/fp16/decorators.py
code served (permissive licence) · get_code("c610293ce05cf52a")
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 torch
import torch.nn as nn
import functools
from inspect import getfullargspec

torch.manual_seed(0)

# Helper function that's referenced in the decorator
def cast_tensor_type(value, src_type, dst_type):
    """Convert tensor types recursively."""
    if isinstance(value, torch.Tensor):
        if value.dtype == src_type:
            return value.to(dst_type)
        return value
    elif isinstance(value, (list, tuple)):
        return type(value)(cast_tensor_type(v, src_type, dst_type) for v in value)
    elif isinstance(value, dict):
        return {k: cast_tensor_type(v, src_type, dst_type) for k, v in value.items()}
    return value

# Create a minimal nn.Module with fp16_enabled attribute
class TestModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.fp16_enabled = True
    
    @functools.wraps  # placeholder
    def method_to_decorate(self, x, y, z=None):
        """Test method that would be decorated."""
        return x + y

# Create test module instance
test_module = TestModule()

# Create some test tensors (fp16)
x_fp16 = torch.randn(2, 4, dtype=torch.float16)
y_fp16 = torch.randn(2, 4, dtype=torch.float16)

ARGS = {
    'apply_to': ('x', 'y'),
    'out_fp16': 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 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
ga_loc_target Ran
ran on a synthesized input
FayCai/framework/mmdet/core/anchor/guided_anchor_target.py
code served (permissive licence) · get_code("2440313353f0c835")
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 ground truth bboxes for 2 images
# Format: [x1, y1, x2, y2] for each bbox
# Image 1: 2 bboxes
gt_bboxes_img1 = torch.tensor([
    [10.0, 10.0, 50.0, 50.0],
    [60.0, 60.0, 100.0, 100.0]
], dtype=torch.float32)

# Image 2: 1 bbox
gt_bboxes_img2 = torch.tensor([
    [20.0, 20.0, 80.0, 80.0]
], dtype=torch.float32)

gt_bboxes_list = [gt_bboxes_img1, gt_bboxes_img2]

# Feature map sizes at different pyramid levels
# Format: (height, width) for each level
featmap_sizes = [
    (64, 64),   # Level 0
    (32, 32),   # Level 1
    (16, 16),   # Level 2
]

# Anchor configuration
anchor_scale = 4
anchor_strides = [4, 8, 16]  # Stride for each pyramid level

# Center and ignore ratios
center_ratio = 0.2
ignore_ratio = 0.5

ARGS = {
    'gt_bboxes_list': gt_bboxes_list,
    'featmap_sizes': featmap_sizes,
    'anchor_scale': anchor_scale,
    'anchor_strides': anchor_strides,
    'center_ratio': center_ratio,
    'ignore_ratio': ignore_ratio,
}
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
get_3rd_point Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
RuiminChen/centernet/mmdet/models/detectors/centernet.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from kmax2001/EvSharp2Blur, is pointer only (licence: NONE) · get_code("9084d28f30c5495b")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from kmax2001/EvSharp2Blur, 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.

get_class_name Ran
ran on a synthesized input
hSDB-instrument/hsdb-mmdetection/hsdb_demo/demo_inference.py
code served (permissive licence) · get_code("f5c79a4971916120")
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 torch

torch.manual_seed(0)

# The function takes:
# - lbl: an integer label
# - class_names: a list/sequence of class name strings

# Reading the function body:
# - lbl is checked against range(len(class_names))
# - class_names is indexed with lbl if lbl is in valid range
# - Otherwise lbl is formatted into a string

# Set up test arguments
lbl = 1
class_names = ['cat', 'dog', 'bird']

ARGS = {
    'lbl': lbl,
    'class_names': class_names
}
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 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_dir Ran
ran, violated the contract we drafted
RuiminChen/centernet/mmdet/models/detectors/centernet.py
code served (permissive licence) · get_code("5d546e1ecbee1866")
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 float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Analyze the function:
# - src_point: used with indexing [0] and [1], so it needs to be indexable with 2 elements
# - rot_rad: used with np.sin() and np.cos(), so it should be a scalar number

# Create test arguments
src_point = [1.0, 2.0]  # A simple 2D point
rot_rad = np.pi / 4     # A rotation angle in radians

ARGS = {
    "src_point": src_point,
    "rot_rad": rot_rad
}
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 float — 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_final_epoch Ran
ran, honoured the contract we drafted
13952522076/mmdet0/.dev_scripts/gather_models.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from mickeykang16/Ev-Stereo3D, is pointer only (licence: NONE) · get_code("e86ae414e2afc197")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from mickeykang16/Ev-Stereo3D, 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.

get_final_epoch Ran
ran, honoured the contract we drafted
ZZR8066/mmdetection-GaofenCompetition/.dev_scripts/gather_models.py
code served (permissive licence) · get_code("a7222b12f224f0ad")
How it ran

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

ARGS = {
    'config': 'resnet50_grid_rcnn_2x_config'
}
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 int, ndim 0, dtype family int, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_final_results Ran
ran; our contract draft was wrong, not the code
13952522076/mmdet0/.dev_scripts/gather_models.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from mickeykang16/Ev-Stereo3D, is pointer only (licence: NONE) · get_code("fee6fb3d147348e3")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from mickeykang16/Ev-Stereo3D, 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.

get_final_results Ran
ran; our contract draft was wrong, not the code
JHuang-CV/mmdetectionV2/.dev_scripts/gather_models.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from zhaoyangwei123/sapnet, is pointer only (licence: NONE) · get_code("2268355e37eebb82")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from zhaoyangwei123/sapnet, 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.

get_p2l_dis Ran
ran on a synthesized input
abcxs/polyrnn/approx_poly.py
code served (permissive licence) · get_code("ff4e60a289af3fb1")
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 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 math

def get_p2l_dis(point_x: int, point_y: int, line_x1: int, line_y1: int, line_x2: int, line_y2: int) -> float:
    """
    【计算点P到直线L的最短距离】
    :param point_x: P的X坐标,int(在本工程中出现的所有坐标均采用整数型)
    :param point_y: P的Y坐标
    :param line_x1: 直线上的点L1的X坐标
    :param line_y1: 直线上的点L1的Y坐标
    :param line_x2: 直线上的点L2的X坐标
    :param line_y2: 直线上的点L2的Y坐标
    :return: 距离值,float
    """
    va = line_y2 - line_y1
    vb = line_x1 - line_x2
    vc = line_x2 * line_y1 - line_x1 * line_y2
    dis = math.fabs(va * point_x + vb * point_y + vc) / math.pow(va * va + vb * vb, 0.5)
    return dis


ARGS = {
    "point_x": 5,
    "point_y": 3,
    "line_x1": 0,
    "line_y1": 0,
    "line_x2": 10,
    "line_y2": 10,
}
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 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
get_training_augmentation Ran
ran on a synthesized input
sathishkumark27/mmdetection-icevision/albu/augmentations.py
code served (permissive licence) · get_code("ba9fcf991899c6ba")
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 Compose. 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 albumentations as albu

ARGS = {
    'min_area': 0.0,
    'min_visibility': 0.0,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
albumentations — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Compose — a Compose 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_validation_augmentation Ran
ran on a synthesized input
sathishkumark27/mmdetection-icevision/albu/augmentations.py
code served (permissive licence) · get_code("d99a525c8bd2f445")
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 Compose. 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 albumentations as albu

ARGS = {
    'min_area': 0.0,
    'min_visibility': 0.0,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
albumentations — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Compose — a Compose output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Compose 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
head_filter Ran
ran on a synthesized input
floatingstarZ/loc_cls_exp/CONTROL_CODE/config.py
code served (permissive licence) · get_code("22e752595ce826e7")
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 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)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.head = nn.Linear(4, 3)
        self.body = nn.Linear(3, 2)
        self.head_layer2 = nn.Linear(2, 1)
        self.other = nn.Linear(1, 1)
    
model = SimpleModel()

ARGS = {
    'model': model
}
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 1, dtype family float — 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
images_to_levels Ran
ran; our contract draft was wrong, not the code
HAOCHENYE/yehc_mmdet/mmdet/core/anchor/utils.py
code served (permissive licence) · get_code("0910773f52297bf8")
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 3, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Analyze the function:
# - target: a list/sequence that gets stacked into a tensor with torch.stack
# - num_levels: a sequence of integers representing the number of levels and their sizes
# 
# The function stacks targets by image, then splits them by levels.
# target should be a list of tensors (one per image)
# num_levels should be a sequence of integers that sum to the feature dimension

# Create a list of target tensors (one per image)
# Each tensor represents features for that image
# Let's say we have 2 images, and features of size 4
target = [
    torch.randn(4),  # image 0: 4 features
    torch.randn(4),  # image 1: 4 features
]

# num_levels specifies how to split the 4 features into levels
# For example: [2, 2] means level 0 gets first 2 features, level 1 gets next 2
num_levels = [2, 2]

ARGS = {
    "target": target,
    "num_levels": num_levels,
}
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 3, dtype family float — 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
images_to_levels Ran
ran on a synthesized input
FayCai/framework/mmdet/core/anchor/anchor_target.py
code served (permissive licence) · get_code("8fa249eba3a0ce57")
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

torch.manual_seed(0)

# The function expects:
# - target: a list of tensors (one per image)
# - num_level_anchors: a list of integers representing anchor counts per level

# Create a list of target tensors (one per image)
# Each tensor has shape (num_anchors,) where num_anchors is the total across all levels
target = [
    torch.randn(7),  # Image 0: 7 total anchors
    torch.randn(7),  # Image 1: 7 total anchors
]

# Define how many anchors per level
# Total anchors = 3 + 4 = 7
num_level_anchors = [3, 4]

ARGS = {
    "target": target,
    "num_level_anchors": num_level_anchors,
}
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 — 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
images_to_levels Ran
ran on a synthesized input
silist/mmdet_dev/mmdet/core/anchor/point_target.py
code served (permissive licence) · get_code("99bbbfc95abdc86a")
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

torch.manual_seed(0)

# Analyzing the function:
# - target: a list of tensors (one per image) that will be stacked into shape (num_images, total_grid_points)
# - num_level_grids: a list of integers representing grid counts per level
# 
# The function stacks targets, then splits them by level based on num_level_grids

# Create target list - 2 images with features at different grid points
# Total grid points should equal sum of num_level_grids
num_level_grids = [6, 4, 2]  # 3 levels with 6, 4, 2 grid points respectively
total_grid_points = sum(num_level_grids)  # 12

# Create 2 image targets, each with 12 grid point features
target = [
    torch.randn(total_grid_points),  # image 0
    torch.randn(total_grid_points),  # image 1
]

ARGS = {
    "target": target,
    "num_level_grids": num_level_grids,
}
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 — 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
iou Ran
ran, violated the contract we drafted
mengfu188/mmdetection.bak/sort.py
code served (permissive licence) · get_code("d84d92dcfe0933de")
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 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 numpy as np
from numba import jit

@jit
def iou(bb_test, bb_gt):
    """
    Computes IOU between two bboxes in the form [x1,y1,x2,y2]
    """
    xx1 = np.maximum(bb_test[0], bb_gt[0])
    yy1 = np.maximum(bb_test[1], bb_gt[1])
    xx2 = np.minimum(bb_test[2], bb_gt[2])
    yy2 = np.minimum(bb_test[3], bb_gt[3])
    w = np.maximum(0.0, xx2 - xx1)
    h = np.maximum(0.0, yy2 - yy1)
    wh = w * h
    o = wh / ((bb_test[2] - bb_test[0]) * (bb_test[3] - bb_test[1]) + (bb_gt[2] - bb_gt[0]) * (bb_gt[3] - bb_gt[1]) - wh)
    return o

bb_test = np.array([10.0, 20.0, 50.0, 80.0])
bb_gt = np.array([15.0, 25.0, 60.0, 90.0])

ARGS = {
    "bb_test": bb_test,
    "bb_gt": bb_gt,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numba, numpy — 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
parse_args Ran
ran on a synthesized input
floatingstarZ/loc_cls_exp/train_for_crowd_human.py
code served (permissive licence) · get_code("5e9fad90ce01f91a")
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 Namespace. 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 argparse
import os
import tempfile

# Create a temporary config file path (won't actually be read by parse_args)
temp_dir = tempfile.gettempdir()
config_path = os.path.join(temp_dir, 'dummy_config.py')

ARGS = {
    'args': [
        config_path,
        '--work_dir', os.path.join(temp_dir, 'work_dir'),
        '--resume_from', os.path.join(temp_dir, 'checkpoint.pth'),
        '--validate',
        '--gpus', '2',
        '--seed', '42',
        '--launcher', 'none',
        '--local_rank', '0',
        '--autoscale-lr'
    ]
}
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 Namespace — a Namespace output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Namespace 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
patch_forward_method Ran
ran; our contract draft was wrong, not the code
nithinksath96/MMdetection_TensorRT_FP16/mmdet/core/fp16/hooks.py
code served (permissive licence) · get_code("e059fcc920f08632")
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 torch
import torch.nn as nn

torch.manual_seed(0)

# Helper function that the patched method will use
def cast_tensor_type(obj, src_type, dst_type):
    """Cast tensors from src_type to dst_type."""
    if isinstance(obj, torch.Tensor):
        return obj.to(dst_type)
    elif isinstance(obj, (tuple, list)):
        return type(obj)(cast_tensor_type(item, src_type, dst_type) for item in obj)
    elif isinstance(obj, dict):
        return {k: cast_tensor_type(v, src_type, dst_type) for k, v in obj.items()}
    return obj

# Create a simple forward method that takes tensors and returns tensors
def sample_forward(x, y=None):
    """Sample forward method that processes tensors."""
    result = x + 1.0
    if y is not None:
        result = result + y
    return result

# Create test tensors with source type (float32) and destination type (float64)
src_dtype = torch.float32
dst_dtype = torch.float64

# Create sample input tensors
x_tensor = torch.randn(2, 4, dtype=src_dtype)
y_tensor = torch.randn(2, 4, dtype=src_dtype)

ARGS = {
    "func": sample_forward,
    "src_type": src_dtype,
    "dst_type": dst_dtype,
    "convert_output": 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 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
patch_norm_fp32 Ran
ran; our contract draft was wrong, not the code
nithinksath96/MMdetection_TensorRT_FP16/mmdet/core/fp16/hooks.py
code served (permissive licence) · get_code("289c48a77b483795")
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 SimpleModel. 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 small module with BatchNorm and GroupNorm layers
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.bn = nn.BatchNorm2d(4)
        self.gn = nn.GroupNorm(2, 4)
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        x = self.bn(x)
        x = self.gn(x)
        return x

module = SimpleModel()

ARGS = {
    "module": module,
}
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 SimpleModel — a SimpleModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleModel 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
polygon_fitting Ran
ran on a synthesized input
abcxs/polyrnn/approx_poly.py
code served (permissive licence) · get_code("10686c0472030409")
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 int, shape (4, 1, 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 numpy as np
import cv2

# Create a simple contour point set
# The function expects an n*1*2 ndarray of points
points = np.array([
    [[10, 20]],
    [[15, 25]],
    [[20, 22]],
    [[25, 30]],
    [[30, 28]],
    [[35, 32]],
    [[40, 30]],
    [[38, 20]],
    [[30, 15]],
    [[20, 18]],
], dtype=np.int32)

# Create a small test image (100x100 RGB image)
image = np.ones((100, 100, 3), dtype=np.uint8) * 255

ARGS = {
    'points': points,
    'delta': 5.0,
    'image': image,
    'times': 2,
    'flag_image': False,
    'flag_text': True,
    'flag_debug': False,
    'color_point': (0, 0, 255),
    'color_line': (0, 255, 0),
    'color_text': (0, 127, 255),
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, 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 int, shape (4, 1, 2) — shape (4, 1, 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
polygon_fitting_main Ran
ran on a synthesized input
abcxs/polyrnn/approx_poly.py
code served (permissive licence) · get_code("ef5996b22e1fea90")
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 (6, 1, 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 numpy as np
import math

# Create a small polygon contour point set
# The function expects n*1*2 ndarray format
# Let's create a simple polygon with 10 points
points = np.array([
    [[0, 0]],
    [[1, 1]],
    [[2, 1]],
    [[3, 2]],
    [[4, 2]],
    [[5, 1]],
    [[6, 0]],
    [[5, -1]],
    [[3, -1]],
    [[1, -1]]
], dtype=float)

# Delta threshold for polygon fitting
delta = 0.5

# Debug flag
flag_debug = False

ARGS = {
    "points": points,
    "delta": delta,
    "flag_debug": flag_debug
}
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 (6, 1, 2) — shape (6, 1, 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
process_checkpoint Ran
ran; our contract draft was wrong, not the code
13952522076/mmdet0/.dev_scripts/gather_models.py
code served (permissive licence) · get_code("e76a491553cf9fcf")
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 torch
import tempfile
import os

torch.manual_seed(0)

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

# Create a minimal checkpoint dict with a model state and optimizer
checkpoint = {
    'model': {
        'layer1.weight': torch.randn(4, 3),
        'layer1.bias': torch.randn(4),
    },
    'optimizer': {
        'state': {},
        'param_groups': [],
    },
    'epoch': 1,
}

# Save the checkpoint to a temporary file
in_file = os.path.join(temp_dir, 'checkpoint.pth')
torch.save(checkpoint, in_file)

# Define output file path
out_file = os.path.join(temp_dir, 'checkpoint_processed.pth')

ARGS = {
    'in_file': in_file,
    'out_file': out_file,
}
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 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
process_checkpoint Ran
ran; our contract draft was wrong, not the code
fehmikahraman/mmdetection/.dev_scripts/gather_models.py
code served (permissive licence) · get_code("138e112e22b0314e")
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 torch
import tempfile
import os

torch.manual_seed(0)

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

# Create a valid checkpoint file
in_file = os.path.join(tmpdir, 'checkpoint.pth')
out_file = os.path.join(tmpdir, 'processed.pth')

# Build a checkpoint with state_dict, optimizer, and ema_ keys
checkpoint = {
    'state_dict': {
        'layer1.weight': torch.randn(4, 3),
        'layer1.bias': torch.randn(4),
        'ema_layer1.weight': torch.randn(4, 3),
        'ema_layer1.bias': torch.randn(4),
    },
    'optimizer': {
        'state': {},
        'param_groups': []
    },
    'epoch': 10
}

# Save the checkpoint to disk
torch.save(checkpoint, in_file)

ARGS = {
    'in_file': in_file,
    'out_file': out_file
}
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 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
process_checkpoint Ran
ran; our contract draft was wrong, not the code
jichengYUAN/mmdetectionCust/.dev_scripts/gather_models.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from zhaoyangwei123/sapnet, is pointer only (licence: NONE) · get_code("ae35528c6bd4bd1c")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from zhaoyangwei123/sapnet, 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.

process_model_info Ran
ran; our contract draft was wrong, not the code
fehmikahraman/mmdetection/.dev_scripts/convert_test_benchmark_script.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from opendatalab/MLS-BRN, is pointer only (licence: NONE) · get_code("893e052530d28946")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from opendatalab/MLS-BRN, 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.

random_boxes Ran
ran, honoured the contract we drafted
HAOCHENYE/yehc_mmdet/mmdet/core/bbox/demodata.py
code served (permissive licence) · get_code("00430ee7a62da222")
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 numpy as np
import torch

torch.manual_seed(0)

ARGS = {
    "num": 2,
    "scale": 512,
    "rng": 42,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
random_boxes Ran
ran, honoured the contract we drafted
dereyly/mmdet_sota/mmdet/core/bbox/demodata.py
code served (permissive licence) · get_code("3dc30523833b925a")
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, 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
import torch

torch.manual_seed(0)

ARGS = {
    "num": 5,
    "scale": 512,
    "rng": 42,
}
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, 4) — shape (5, 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
read_tsv Ran
ran on a synthesized input
sathishkumark27/mmdetection-icevision/albu/augmenter.py
code served (permissive licence) · get_code("d659ebd4df6fa2be")
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 DataFrame. 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 pandas as pd
import io

# Create a minimal TSV file content as a string
tsv_content = "col1\tcol2\tcol3\n1\t2\t3\n4\t5\t6\n7\t8\t9"

# Create a file-like object from the string
file = io.StringIO(tsv_content)

ARGS = {
    "file": file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 and dtype family int 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
show_det_result Ran
ran on a synthesized input
abcxs/polyrnn/demo_server/model.py
code served (permissive licence) · get_code("d69c73a87e135977")
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 int, shape (100, 100, 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 cv2

# Create a small test image
img = np.ones((100, 100, 3), dtype=np.uint8) * 255

# Create detection results
# bbox_result should be a list of arrays, each array contains bboxes with format [x1, y1, x2, y2, score]
# Creating 2 detection results (e.g., for 2 classes)
bbox_result_class1 = np.array([
    [10, 10, 50, 50, 0.9],
    [60, 60, 90, 90, 0.7],
], dtype=np.float32)

bbox_result_class2 = np.array([
    [20, 20, 40, 40, 0.85],
], dtype=np.float32)

result = [bbox_result_class1, bbox_result_class2]

# Score threshold
score_thr = 0.75

ARGS = {
    'img': img,
    'result': result,
    'score_thr': score_thr,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, 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 int, shape (100, 100, 3) — shape (100, 100, 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
show_polygon_result Ran
ran on a synthesized input
abcxs/polyrnn/demo_server/model.py
code served (permissive licence) · get_code("e72b842cae1bef0b")
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 cv2

# Create a small test image (100x100 RGB)
img = np.zeros((100, 100, 3), dtype=np.uint8)

# Create bbox_result: list of arrays with shape (N, 5) where last column is score
# Each bbox has [x1, y1, x2, y2, score]
bbox_array_1 = np.array([
    [10.0, 10.0, 30.0, 30.0, 0.9],
    [40.0, 40.0, 60.0, 60.0, 0.7],
    [70.0, 70.0, 90.0, 90.0, 0.3],
], dtype=np.float32)

bbox_result = [bbox_array_1]

# Create polygon_result: list containing list of polygons
# Each polygon is a flat list of coordinates [x1, y1, x2, y2, x3, y3, ...]
polygon_result_1 = [
    [12.0, 12.0, 28.0, 12.0, 28.0, 28.0, 12.0, 28.0],  # 4 points for bbox 0
    [42.0, 42.0, 58.0, 42.0, 58.0, 58.0, 42.0, 58.0],  # 4 points for bbox 1
    [72.0, 72.0, 88.0, 72.0, 88.0, 88.0, 72.0, 88.0],  # 4 points for bbox 2
]

polygon_result = [polygon_result_1]

# Create result tuple: (bbox_result, _, polygon_result)
result = (bbox_result, None, polygon_result)

# score_thr parameter
score_thr = 0.5

ARGS = {
    'img': img,
    'result': result,
    'score_thr': score_thr,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, 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
snow_process Ran
ran on a synthesized input
sathishkumark27/mmdetection-icevision/augmentations/augmentations.py
code served (permissive licence) · get_code("e483fd8843430f6f")
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 int, 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 cv2
import numpy as np

# Create a small test image in RGB format
# Using shape (4, 3, 3) for height=4, width=3, channels=3
image = np.random.randint(0, 256, size=(4, 3, 3), dtype=np.uint8)

# snow_coeff is a threshold value for the brightness channel
snow_coeff = 100

ARGS = {
    "image": image,
    "snow_coeff": snow_coeff
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, 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 int, 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
transform_preds Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
RuiminChen/centernet/mmdet/models/detectors/centernet.py
code served (permissive licence) · get_code("a22a1ae3bbb508ce")
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, 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 cv2
import numpy as np
import torch

torch.manual_seed(0)

# Define small test data
coords = np.array([
    [10.5, 20.3, 0.0],
    [15.2, 25.8, 0.0]
], dtype=np.float32)

center = np.array([64.0, 64.0], dtype=np.float32)
scale = np.array([128.0, 128.0], dtype=np.float32)
output_size = (256, 256)

ARGS = {
    'coords': coords,
    'center': center,
    'scale': scale,
    'output_size': output_size
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, 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, 3) — shape (2, 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
xcorr_depthwise Ran
ran on a synthesized input
YeLyuUT/MMDET/mmdet/core/xcorr.py
code served (permissive licence) · get_code("4cb416b55d390cbd")
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, 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 torch.nn.functional as F

torch.manual_seed(0)

# Construct tensors for xcorr_depthwise
# The function expects:
# - x: a 4D tensor that will be reshaped to (1, batch*channel, height, width)
# - kernel: a 4D tensor of shape (batch, channel, kernel_h, kernel_w)
# - padding: an integer (default 0)

batch = 2
channel = 3
height = 5
width = 6
kernel_h = 3
kernel_w = 3

# Create kernel tensor: (batch, channel, kernel_h, kernel_w)
kernel = torch.randn(batch, channel, kernel_h, kernel_w)

# Create x tensor: needs to have compatible spatial dimensions
# After conv2d with kernel_size=3 and padding=0, output will be smaller
# x should be (batch, channel, height, width) based on how it's used
x = torch.randn(batch, channel, height, width)

# padding parameter
padding = 0

ARGS = {
    "x": x,
    "kernel": kernel,
    "padding": padding
}
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, 4) — shape (2, 3, 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

Showing the first 60. An agent gets every one from get_harvested_code_for_paper("1906.07155").

Also present in other papers' code

Of 82 samples lifted from this paper's linked repositories, 82 carry a function identity and 0 do not. Of the 74 distinct functions with an identity, 45 are also present in other papers whose code we hold (16 with at least one paper dated earlier, 0 the same month, 44 later) and 29 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
process_checkpoint 45 other papers
1 dated earlier, 43 later, 1 undated
1904.11490 (2019-04)
RepPoints: Point Set Representation for Object Detection
1908.05612 (2019-08)
R3Det: Refined Single-Stage Detector with Feature Refinement for Rotating Object
1909.04972 (2019-09)
WSOD^2: Learning Bottom-up and Top-down Objectness Distillation for Weakly-supervised Object Detection
and 42 more not named here
get_final_epoch 35 other papers
1 dated earlier, 33 later, 1 undated
1904.11490 (2019-04)
RepPoints: Point Set Representation for Object Detection
1908.05612 (2019-08)
R3Det: Refined Single-Stage Detector with Feature Refinement for Rotating Object
1909.04972 (2019-09)
WSOD^2: Learning Bottom-up and Top-down Objectness Distillation for Weakly-supervised Object Detection
and 32 more not named here
ordered_yaml_dump 31 other papers
1 dated earlier, 30 later
1611.02644 (2016-11)
Multispectral Deep Neural Networks for Pedestrian Detection
2105.01928 (2021-05)
Instances as Queries
2107.00420 (2021-07)
CBNet: A Composite Backbone Network Architecture for Object Detection
and 28 more not named here
process_checkpoint 22 other papers
1 dated earlier, 21 later
1611.02644 (2016-11)
Multispectral Deep Neural Networks for Pedestrian Detection
2105.01928 (2021-05)
Instances as Queries
2108.10520 (2021-08)
Improving Object Detection by Label Assignment Distillation
and 19 more not named here
channel_shuffle 19 other papers
5 dated earlier, 14 later · 2 only as an assembly brick
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition
1606.02147 (2016-06)
ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation
1805.12549 (2018-05)
Channel Gating Neural Networks
and 16 more not named here
get_final_results 10 other papers
10 later
1911.10683 (2019-11)
Image-based table recognition: data, model, and evaluation
2008.13367 (2020-08)
VarifocalNet: An IoU-aware Dense Object Detector
2106.04263 (2021-06)
On the Connection between Local Attention and Dynamic Depth-wise Convolution
and 7 more not named here
process_model_info 10 other papers
10 later
2105.01928 (2021-05)
Instances as Queries
2107.00420 (2021-07)
CBNet: A Composite Backbone Network Architecture for Object Detection
2203.10593 (2022-03)
Open-Vocabulary One-Stage Detection with Hierarchical Visual-Language Knowledge Distillation
and 7 more not named here
check_link 9 other papers
9 later
2204.02136 (2022-04)
Overcoming Catastrophic Forgetting in Incremental Object Detection via Elastic Response Distillation
2209.09841 (2022-09)
Exploring Inconsistent Knowledge Distillation for Object Detection with Data Augmentation
2212.07784 (2022-12)
RTMDet: An Empirical Study of Designing Real-Time Object Detectors
and 6 more not named here
check_path 9 other papers
9 later
2204.02136 (2022-04)
Overcoming Catastrophic Forgetting in Incremental Object Detection via Elastic Response Distillation
2209.09841 (2022-09)
Exploring Inconsistent Knowledge Distillation for Object Detection with Data Augmentation
2212.07784 (2022-12)
RTMDet: An Empirical Study of Designing Real-Time Object Detectors
and 6 more not named here
check_url 9 other papers
9 later
2204.02136 (2022-04)
Overcoming Catastrophic Forgetting in Incremental Object Detection via Elastic Response Distillation
2209.09841 (2022-09)
Exploring Inconsistent Knowledge Distillation for Object Detection with Data Augmentation
2212.07784 (2022-12)
RTMDet: An Empirical Study of Designing Real-Time Object Detectors
and 6 more not named here
get_final_results 8 other papers
1 dated earlier, 7 later
1904.11490 (2019-04)
RepPoints: Point Set Representation for Object Detection
1908.05612 (2019-08)
R3Det: Refined Single-Stage Detector with Feature Refinement for Rotating Object
1909.04972 (2019-09)
WSOD^2: Learning Bottom-up and Top-down Objectness Distillation for Weakly-supervised Object Detection
and 5 more not named here
get_dir 7 other papers
7 later · 1 only as an assembly brick
1908.07919 (2019-08)
Deep High-Resolution Representation Learning for Visual Recognition
2105.04154 (2021-05)
Unsupervised Human Pose Estimation through Transforming Shape Templates
2107.13774 (2021-07)
Geometry Uncertainty Projection Network for Monocular 3D Object Detection
and 4 more not named here
process_checkpoint 7 other papers
7 later
2107.00420 (2021-07)
CBNet: A Composite Backbone Network Architecture for Object Detection
2108.07755 (2021-08)
TOOD: Task-aligned One-stage Object Detection
2110.13389 (2021-10)
A Normalized Gaussian Wasserstein Distance for Tiny Object Detection
and 4 more not named here
images_to_levels 6 other papers
1 dated earlier, 5 later
1903.00621 (2019-03)
Feature Selective Anchor-Free Module for Single-Shot Object Detection
1907.03958 (2019-07)
Attentive CT Lesion Detection Using Deep Pyramid Inference with Multi-Scale Booster
2003.10152 (2020-03)
SOLOv2: Dynamic and Fast Instance Segmentation
and 3 more not named here
get_3rd_point 5 other papers
5 later
1908.07919 (2019-08)
Deep High-Resolution Representation Learning for Visual Recognition
2105.04154 (2021-05)
Unsupervised Human Pose Estimation through Transforming Shape Templates
2403.08019 (2024-03)
MRC-Net: 6-DoF Pose Estimation with MultiScale Residual Correlation
and 2 more not named here
anchor_target 4 other papers
1 dated earlier, 3 later
1903.00621 (2019-03)
Feature Selective Anchor-Free Module for Single-Shot Object Detection
2003.10152 (2020-03)
SOLOv2: Dynamic and Fast Instance Segmentation
2004.06002 (2020-04)
Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training
and 1 more not named here
calc_region 4 other papers
1 dated earlier, 3 later
1903.00621 (2019-03)
Feature Selective Anchor-Free Module for Single-Shot Object Detection
2003.10152 (2020-03)
SOLOv2: Dynamic and Fast Instance Segmentation
2004.06002 (2020-04)
Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training
and 1 more not named here
convert_bbox_to_z 4 other papers
3 dated earlier, 1 later
1602.00763 (2016-02)
Simple Online and Realtime Tracking
1612.01465 (2016-12)
ArtTrack: Articulated Multi-person Tracking in the Wild
and 1 more not named here
convert_x_to_bbox 4 other papers
3 dated earlier, 1 later
1602.00763 (2016-02)
Simple Online and Realtime Tracking
1612.01465 (2016-12)
ArtTrack: Articulated Multi-person Tracking in the Wild
and 1 more not named here
ga_loc_target 4 other papers
1 dated earlier, 3 later
1903.00621 (2019-03)
Feature Selective Anchor-Free Module for Single-Shot Object Detection
2003.10152 (2020-03)
SOLOv2: Dynamic and Fast Instance Segmentation
2004.06002 (2020-04)
Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training
and 1 more not named here

Shown: the 20 functions present in the most other papers. 25 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

We present MMDetection, an object detection toolbox that contains a rich set of object detection and instance segmentation methods as well as related components and modules. The toolbox started from a codebase of MMDet team who won the detection track of COCO Challenge 2018. It gradually evolves into a unified platform that covers many popular detection methods and contemporary modules. It not only includes training and inference codes, but also provides weights for more than 200 network models. We believe this toolbox is by far the most complete detection toolbox. In this paper, we introduce the various features of this toolbox. In addition, we also conduct a benchmarking study on different methods, components, and their hyper-parameters. We wish that the toolbox and benchmark could serve the growing research community by providing a flexible toolkit to reimplement existing methods and develop their own new detectors. Code and models are available at https://github.com/open-mmlab/mmdetection. The project is under active development and we will keep this document updated.

For agents

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

get_harvested_code_for_paper("1906.07155")
get_code_for_paper("1906.07155")
have("1906.07155")

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

curl https://syntology.ai/api/ran/1906.07155.json

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

[![Syntology run record](https://syntology.ai/api/ran/1906.07155.svg)](https://syntology.ai/paper/1906.07155)

Connect an agent — have() is free.