SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2204.00826 · CVPR · 2022

Online Convolutional Re-parameterization

Jianqiang Huang, Hua, B Relu, Baisheng Lai, Xiaojin Gong, Mu Hu, Junyi Feng, Jiashen Hua

arXiv · PDF · Open in the Atlas

Code that ran

5 of 6 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 3 have no contract verdict; 2 failures of our instrument; 1 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/2204.00826.json · badge: /api/ran/2204.00826.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
JUGGHM/OREPA_CVPR2022 canonical (classifier label)
classifier basis not recorded on the link
5 ran · 1 no recorded run
FunctionStatusWhere it lives
accuracy Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
JUGGHM/OREPA_CVPR2022/utils.py
code served (permissive licence) · get_code("b0f936d4d6ae3b8c")
How it ran

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

import torch

torch.manual_seed(0)

# Create output tensor: batch_size=2, num_classes=4
# This represents logits from a model
output = torch.randn(2, 4)

# Create target tensor: batch_size=2
# These are class indices (0-3 for 4 classes)
target = torch.tensor([1, 3])

# topk parameter: tuple of k values to compute accuracy for
topk = (1, 2)

ARGS = {
    "output": output,
    "target": target,
    "topk": topk,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 2, 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_ImageNet_train_dataset Ran
ran on a synthesized input
JUGGHM/OREPA_CVPR2022/utils.py
code served (permissive licence) · get_code("db9ff684dc155bc8")
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 ImageFolder. 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 torchvision.transforms as transforms
from torchvision.datasets import ImageFolder
import tempfile
import os
from PIL import Image
import shutil

# Create a temporary directory structure that mimics ImageNet train layout
temp_dir = tempfile.mkdtemp()
data_dir = os.path.join(temp_dir, 'data')
train_dir = os.path.join(data_dir, 'train')
os.makedirs(train_dir, exist_ok=True)

# Create a few dummy class directories with dummy images
for class_idx in range(2):
    class_dir = os.path.join(train_dir, f'class_{class_idx}')
    os.makedirs(class_dir, exist_ok=True)
    
    # Create 2 dummy images per class
    for img_idx in range(2):
        img = Image.new('RGB', (32, 32), color=(73, 109, 137))
        img_path = os.path.join(class_dir, f'image_{img_idx}.jpg')
        img.save(img_path)

# Define transform
trans = transforms.Compose([
    transforms.Resize((224, 224)),
    transforms.ToTensor(),
])

# Create args object
class Args:
    def __init__(self, data_path):
        self.data = data_path

args = Args(data_dir)

ARGS = {
    'args': args,
    'trans': trans,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ImageFolder — a ImageFolder output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
read_hdf5 Ran
ran on a synthesized input
JUGGHM/OREPA_CVPR2022/utils.py
code served (permissive licence) · get_code("772b15eb299c691c")
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 h5py
import numpy as np
import tempfile
import os

# Create a temporary HDF5 file with test data
temp_dir = tempfile.mkdtemp()
test_file = os.path.join(temp_dir, 'test_data.h5')

with h5py.File(test_file, 'w') as f:
    f.create_dataset('array1', data=np.array([[1, 2, 3, 4], [5, 6, 7, 8]]))
    f.create_dataset('array2', data=np.array([9, 10, 11, 12]))
    f.create_dataset('nested+group', data=np.array([[[1, 2], [3, 4]], [[5, 6], [7, 8]]]))

ARGS = {
    'file_path': test_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
h5py, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
transII_addbranch Ran
ran on a synthesized input
JUGGHM/OREPA_CVPR2022/dbb_transforms.py
code served (permissive licence) · get_code("788ed0b33aadf578")
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 torch.nn as nn

torch.manual_seed(0)

# The function sums kernels and sums biases
# kernels and biases should be iterables (lists, tuples) of tensors
# that support sum() operation

kernels = [
    torch.randn(4, 3),
    torch.randn(4, 3),
    torch.randn(4, 3)
]

biases = [
    torch.randn(2),
    torch.randn(2),
    torch.randn(2)
]

ARGS = {
    "kernels": kernels,
    "biases": biases
}
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
transI_fusebn Ran
ran; our contract draft was wrong, not the code
JUGGHM/OREPA_CVPR2022/dbb_transforms.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from SCUT-DLVCLab/AutoHDR, is pointer only (licence: NONE) · get_code("079d0a85a0bfbb47")
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 SCUT-DLVCLab/AutoHDR, 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.

transIII_1x1_kxk Not yet run
unverified (harvested, no recorded run)
JUGGHM/OREPA_CVPR2022/dbb_transforms.py
code served (permissive licence) · get_code("a1bfd1b41b138e66")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 3 later) and 2 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
accuracy 116 other papers
41 dated earlier, 71 later, 4 same year, order not determinable · 8 only as an assembly brick
1502.03044 (2015-02)
Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
1610.05820 (2016-10)
Membership Inference Attacks Against Machine Learning Models
1905.03696 (2019-05)
HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precision
and 113 more not named here
transI_fusebn 5 other papers
1 dated earlier, 4 later
2109.03508 (2021-09)
RepNAS: Searching for Efficient Re-parameterizing Blocks
2206.14651 (2022-06)
BoT-SORT: Robust Associations Multi-Pedestrian Tracking
2207.02696 (2022-07)
YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors
and 2 more not named here
transIII_1x1_kxk 2 other papers
1 dated earlier, 1 later
2109.03508 (2021-09)
RepNAS: Searching for Efficient Re-parameterizing Blocks
2312.12470 (2023-12)
Rotated Multi-Scale Interaction Network for Referring Remote Sensing Image Segmentation
transII_addbranch 1 other paper
1 dated earlier
2109.03508 (2021-09)
RepNAS: Searching for Efficient Re-parameterizing Blocks

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

Structural re-parameterization has drawn increasing attention in various computer vision tasks. It aims at improving the performance of deep models without introducing any inference-time cost. Though efficient during inference, such models rely heavily on the complicated training-time blocks to achieve high accuracy, leading to large extra training cost. In this paper, we present online convolutional reparameterization (OREPA), a two-stage pipeline, aiming to reduce the huge training overhead by squeezing the complex training-time block into a single convolution. To achieve this goal, we introduce a linear scaling layer for better optimizing the online blocks. Assisted with the reduced training cost, we also explore some more effective re-param components. Compared with the state-of-the-art re-param models, OREPA is able to save the training-time memory cost by about 70% and accelerate the training speed by around 2×. Meanwhile, equipped with OREPA, the models outperform previous methods on ImageNet by up to +0.6%. We also conduct experiments on object detection and semantic segmentation and show consistent improvements on the downstream tasks. Codes are available at https: //github.com/JUGGHM/OREPA_CVPR2022.

For agents

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

get_harvested_code_for_paper("2204.00826")
get_code_for_paper("2204.00826")
have("2204.00826")

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

curl https://syntology.ai/api/ran/2204.00826.json

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

[![Syntology run record](https://syntology.ai/api/ran/2204.00826.svg)](https://syntology.ai/paper/2204.00826)

Connect an agent — have() is free.