SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2212.07784 · 2022

RTMDet: An Empirical Study of Designing Real-Time Object Detectors

arXiv · PDF · Open in the Atlas

Code that ran

16 of 20 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 12 have no contract verdict; 4 failures of our instrument; 4 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/2212.07784.json · badge: /api/ran/2212.07784.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
open-mmlab/mmrotate canonical (classifier label)
classifier basis not recorded on the link
8 ran · 2 no recorded run
RangiLyu/mmdetection_test community (classifier label)
classifier basis not recorded on the link
5 ran · 2 no recorded run
yuyi1005/point2rbox-mmrotate community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
FunctionStatusWhere it lives
bbox_flip Ran
ran on a synthesized input
open-mmlab/mmrotate/mmrotate/core/bbox/transforms.py
code served (permissive licence) · get_code("376185dc39be213f")
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, 5) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Create a small tensor of bboxes with shape (2, 5)
# Each row: [x, y, width, height, angle]
bboxes = torch.tensor([
    [10.0, 20.0, 30.0, 40.0, np.pi / 4],
    [50.0, 60.0, 70.0, 80.0, np.pi / 6]
], dtype=torch.float32)

# Image shape as tuple (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
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, 5) — shape (2, 5) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
check_link Ran
ran; our contract draft was wrong, not the code
RangiLyu/mmdetection_test/.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
RangiLyu/mmdetection_test/.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.

determine_gpus Ran
ran on a synthesized input
RangiLyu/mmdetection_test/.dev_scripts/convert_train_benchmark_script.py
code served (permissive licence) · get_code("fdef7359919a4918")
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 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

# The function determine_gpus takes a cfg_name string parameter
# and uses string operations (find, in) to check for patterns
# It returns a tuple of (gpus, gpus_pre_node)

ARGS = {
    'cfg_name': '8x_base_config'
}
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 1, dtype family int — 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
get_cls_results Ran
ran on a synthesized input
open-mmlab/mmrotate/mmrotate/core/evaluation/eval_map.py
code served (permissive licence) · get_code("d5e36bcdd9247df7")
How it ran

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

import torch
import numpy as np

torch.manual_seed(0)

# Construct det_results: list of lists, where each inner list contains detection results for each class
# For class_id=0, we'll have detections; for class_id=1, we'll have detections
# Each detection is an array of [x1, y1, x2, y2, score]
det_results = [
    [
        np.array([[10.0, 20.0, 30.0, 40.0, 0.9], [15.0, 25.0, 35.0, 45.0, 0.8]], dtype=np.float32),
        np.array([[50.0, 60.0, 70.0, 80.0, 0.7]], dtype=np.float32),
    ],
    [
        np.array([[12.0, 22.0, 32.0, 42.0, 0.85]], dtype=np.float32),
        np.array([[52.0, 62.0, 72.0, 82.0, 0.75], [55.0, 65.0, 75.0, 85.0, 0.6]], dtype=np.float32),
    ],
]

# Construct annotations: list of dicts with labels, bboxes, and optionally labels_ignore and bboxes_ignore
annotations = [
    {
        'labels': torch.tensor([0, 1, 0], dtype=torch.long),
        'bboxes': torch.tensor([[10.0, 20.0, 30.0, 40.0, 0.0], 
                                 [50.0, 60.0, 70.0, 80.0, 0.0],
                                 [11.0, 21.0, 31.0, 41.0, 0.0]], dtype=torch.float32),
        'labels_ignore': torch.tensor([1], dtype=torch.long),
        'bboxes_ignore': torch.tensor([[100.0, 100.0, 110.0, 110.0, 0.0]], dtype=torch.float32),
    },
    {
        'labels': torch.tensor([0, 1], dtype=torch.long),
        'bboxes': torch.tensor([[12.0, 22.0, 32.0, 42.0, 0.0],
                                 [52.0, 62.0, 72.0, 82.0, 0.0]], dtype=torch.float32),
    },
]

class_id = 0

ARGS = {
    'det_results': det_results,
    'annotations': annotations,
    'class_id': class_id,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_multiscale_patch Ran
ran on a synthesized input
open-mmlab/mmrotate/mmrotate/core/patch/split.py
code served (permissive licence) · get_code("01cb8ca0622cd942")
How it ran

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

# Analyze the function: get_multiscale_patch
# Parameters:
#   - sizes (list): A list of patch sizes
#   - steps (list): A list of steps to slide patches
#   - ratios (list): Multiscale ratios
# 
# The function uses:
#   - len() on sizes and steps (needs to be equal length)
#   - zip(sizes, steps)
#   - product(size_steps, ratios) 
#   - division operations on size and step values
#   - int() conversion
#   - append() on lists
#
# All parameters are lists of numbers (scalars)

ARGS = {
    'sizes': [32, 64, 128],
    'steps': [8, 16, 32],
    'ratios': [0.5, 1.0, 2.0]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, dtype family int — 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
get_pattern_fill Ran
ran on a synthesized input
yuyi1005/point2rbox-mmrotate/mmrotate/models/task_modules/synthesis_generators/point2rbox_generator.py
code served (permissive licence) · get_code("ede6e43c8f430f7c")
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 (80, 100) 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

# The function get_pattern_fill takes two parameters: w and h
# It creates a numpy array of shape (h, w) and draws lines on it
# Both w and h are converted to int, so they can be int, float, or numeric types

ARGS = {
    "w": 100,
    "h": 80,
}
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 2, dtype family float, shape (80, 100) — shape (80, 100) 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_pattern_line Ran
ran on a synthesized input
yuyi1005/point2rbox-mmrotate/mmrotate/models/task_modules/synthesis_generators/point2rbox_generator.py
code served (permissive licence) · get_code("45a05e5326bdf549")
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 (48, 64) 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)

ARGS = {
    "w": 64,
    "h": 48
}
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 Tensor, ndim 2, dtype family float, shape (48, 64) — shape (48, 64) 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_pattern_rose Ran
ran on a synthesized input
yuyi1005/point2rbox-mmrotate/mmrotate/models/task_modules/synthesis_generators/point2rbox_generator.py
code served (permissive licence) · get_code("1ed78f93e2bc1317")
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 (48, 64) 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)

ARGS = {
    "w": 64,
    "h": 48,
}
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 Tensor, ndim 2, dtype family float, shape (48, 64) — shape (48, 64) 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
map_masks Ran
ran on a synthesized input
open-mmlab/mmrotate/mmrotate/core/patch/merge_results.py
code served (permissive licence) · get_code("101fcc7bda94ac46")
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 numpy as np

# Create small test masks with different sizes
mask1 = np.random.rand(5, 6) > 0.5  # 5 height, 6 width
mask2 = np.random.rand(4, 7) > 0.5  # 4 height, 7 width
masks = [mask1, mask2]

# Create offset as numpy array with shape (2,)
offset = np.array([2, 3])

# Create new_shape as tuple (width, height)
new_shape = (20, 25)

ARGS = {
    "masks": masks,
    "offset": offset,
    "new_shape": new_shape,
}
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 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
process_checkpoint Ran
ran; our contract draft was wrong, not the code
RangiLyu/mmdetection_test/.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_model_info Ran
ran; our contract draft was wrong, not the code
RangiLyu/mmdetection_test/.dev_scripts/convert_test_benchmark_script.py
code served (permissive licence) · get_code("3175ca3b1ec8c7b9")
How it ran

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

import os.path as osp

# Analyze what process_model_info needs:
# - model_info: dict with 'config' and 'checkpoint' keys (both strings with .strip() called)
# - work_dir: string parameter (though it's overwritten in the function)

model_info = {
    'config': '  /path/to/config_file.py  ',  # Has whitespace to test .strip()
    'checkpoint': '  /path/to/checkpoint.pth  '  # Has whitespace to test .strip()
}

work_dir = '/some/work/directory'

ARGS = {
    'model_info': model_info,
    'work_dir': work_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type 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
rbbox2result Ran
ran on a synthesized input
open-mmlab/mmrotate/mmrotate/core/bbox/transforms.py
code served (permissive licence) · get_code("ce792faad2a6fa37")
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

torch.manual_seed(0)

# Create bboxes tensor with shape (n, 6)
# n=2 detections, 6 values per bbox (x, y, w, h, angle, score or similar)
bboxes = torch.tensor([
    [10.0, 20.0, 5.0, 3.0, 0.5, 0.9],
    [30.0, 40.0, 4.0, 2.0, 1.2, 0.8]
], dtype=torch.float32)

# Create labels tensor with shape (n,)
# 2 labels, values in range [0, num_classes-1]
labels = torch.tensor([0, 1], dtype=torch.long)

# Number of classes (including background)
num_classes = 3

ARGS = {
    "bboxes": bboxes,
    "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
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
rotated_anchor_inside_flags Ran
ran on a synthesized input
open-mmlab/mmrotate/mmrotate/core/anchor/utils.py
code served (permissive licence) · get_code("0ea25f06be11d42b")
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)

# flat_anchors: shape (n, 5) - rotated anchors with [cx, cy, w, h, angle]
flat_anchors = torch.tensor([
    [10.0, 15.0, 5.0, 8.0, 0.5],
    [50.0, 60.0, 4.0, 6.0, 1.2],
], 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: (height, width) of the image
img_shape = (100, 120)

# allowed_border: border tolerance in pixels
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
slide_window Ran
ran on a synthesized input
open-mmlab/mmrotate/mmrotate/core/patch/split.py
code served (permissive licence) · get_code("4a5c249cf82b4cba")
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 int, shape (329, 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.

from itertools import product
from math import ceil
import numpy as np

ARGS = {
    'width': 640,
    'height': 480,
    'sizes': [64, 128],
    'steps': [32, 64],
    'img_rate_thr': 0.6,
}
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 int, shape (329, 4) — shape (329, 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
translate_bboxes Ran
ran on a synthesized input
open-mmlab/mmrotate/mmrotate/core/patch/merge_results.py
code served (permissive licence) · get_code("743d0cc9f86f379e")
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, 5) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Case 1: Horizontal bboxes (n, 5) format: (x, y, x, y, score)
bboxes_horizontal = np.array([
    [10.0, 20.0, 30.0, 40.0, 0.9],
    [50.0, 60.0, 70.0, 80.0, 0.8]
], dtype=np.float32)

# Case 2: Rotated bboxes (n, 6) format: (x, y, w, h, theta, score)
bboxes_rotated = np.array([
    [10.0, 20.0, 5.0, 10.0, 0.5, 0.9],
    [50.0, 60.0, 8.0, 12.0, 1.2, 0.8]
], dtype=np.float32)

# Offset to translate with shape (2,)
offset = np.array([5.0, 3.0], dtype=np.float32)

# Use horizontal bboxes as the primary test case
ARGS = {
    "bboxes": bboxes_horizontal,
    "offset": offset
}
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, 5) — shape (2, 5) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
bbox_mapping_back Not yet run
unverified (harvested, no recorded run)
open-mmlab/mmrotate/mmrotate/core/bbox/transforms.py
code served (permissive licence) · get_code("bc4459d2ed8a1d71")
How it ran

no invocation recorded.

check_path Not yet run
unverified (harvested, no recorded run)
RangiLyu/mmdetection_test/.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("fd08138d4bde9cfa")
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. No invocation recorded.

draw_rbboxes Not yet run
unverified (harvested, no recorded run)
open-mmlab/mmrotate/mmrotate/core/visualization/image.py
code served (permissive licence) · get_code("9432da2297676e37")
How it ran

no invocation recorded.

ordered_yaml_dump Not yet run
unverified (harvested, no recorded run)
RangiLyu/mmdetection_test/.dev_scripts/gather_models.py
code served (permissive licence) · get_code("8ff99a6bd2ebf598")
How it ran

no invocation recorded.

Also present in other papers' code

Of 20 samples lifted from this paper's linked repositories, 20 carry a function identity and 0 do not. Of the 20 distinct functions with an identity, 8 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 0 the same month, 8 later) and 12 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
ordered_yaml_dump 31 other papers
14 dated earlier, 17 later
1611.02644 (2016-11)
Multispectral Deep Neural Networks for Pedestrian Detection
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
2105.01928 (2021-05)
Instances as Queries
and 28 more not named here
process_checkpoint 22 other papers
12 dated earlier, 10 later
1611.02644 (2016-11)
Multispectral Deep Neural Networks for Pedestrian Detection
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
2105.01928 (2021-05)
Instances as Queries
and 19 more not named here
check_link 9 other papers
3 dated earlier, 6 later
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
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
and 6 more not named here
check_path 9 other papers
3 dated earlier, 6 later
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
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
and 6 more not named here
check_url 9 other papers
3 dated earlier, 6 later
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
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
and 6 more not named here
process_model_info 3 other papers
1 dated earlier, 2 later
2204.02136 (2022-04)
Overcoming Catastrophic Forgetting in Incremental Object Detection via Elastic Response Distillation
2401.02361 (2024-01)
An Open and Comprehensive Pipeline for Unified Object Grounding and Detection
2505.19028 (2025-05)
InfoChartQA: A Benchmark for Multimodal Question Answering on Infographic Charts
determine_gpus 1 other paper
1 later
2505.19028 (2025-05)
InfoChartQA: A Benchmark for Multimodal Question Answering on Infographic Charts
draw_rbboxes 1 other paper
1 later
aaai_28502 (2024 (year only))
Spatial Transform Decoupling for Oriented Object Detection

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

Repositories linked to this paper

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

Abstract

In this paper, we aim to design an efficient real-time object detector that exceeds the YOLO series and is easily extensible for many object recognition tasks such as instance segmentation and rotated object detection. To obtain a more efficient model architecture, we explore an architecture that has compatible capacities in the backbone and neck, constructed by a basic building block that consists of large-kernel depth-wise convolutions. We further introduce soft labels when calculating matching costs in the dynamic label assignment to improve accuracy. Together with better training techniques, the resulting object detector, named RTMDet, achieves 52.8% AP on COCO with 300+ FPS on an NVIDIA 3090 GPU, outperforming the current mainstream industrial detectors. RTMDet achieves the best parameter-accuracy trade-off with tiny/small/medium/large/extra-large model sizes for various application scenarios, and obtains new state-of-the-art performance on real-time instance segmentation and rotated object detection. We hope the experimental results can provide new insights into designing versatile real-time object detectors for many object recognition tasks. Code and models are released at https://github.com/open-mmlab/mmdetection/tree/3.x/configs/rtmdet.

For agents

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

get_harvested_code_for_paper("2212.07784")
get_code_for_paper("2212.07784")
have("2212.07784")

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

curl https://syntology.ai/api/ran/2212.07784.json

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

[![Syntology run record](https://syntology.ai/api/ran/2212.07784.svg)](https://syntology.ai/paper/2212.07784)

Connect an agent — have() is free.