SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1608.03983 · ICLR · 2016

SGDR: Stochastic Gradient Descent with Warm Restarts

Frank Hutter, Ilya Loshchilov

arXiv · PDF · Open in the Atlas

Code that ran

16 of 17 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (7 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 14 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/1608.03983.json · badge: /api/ran/1608.03983.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
buptlwz/mabp community (classifier label)
classifier basis not recorded on the link
7 ran · 0 no recorded run
jae1213/CyclicalLR_SGDR community
Papers with Code archive
1 ran · 1 no recorded run
Gjiangtao/A-Deep-Supervised-Edge-Optimization-Algorithm-for-Salt-Body-Segmentation community
Papers with Code archive
1 ran · 0 no recorded run
gurucharanmk/PyTorch_CosineAnnealingWithRestartsLR community
Papers with Code archive
1 ran · 0 no recorded run
SevenZhan/Pytorch community
Papers with Code archive
1 ran · 0 no recorded run
Harshvardhan1/cyclic-learning-schedulers-pytorch community
Papers with Code archive
1 ran · 0 no recorded run
bluesky314/Cyclical_LR_Scheduler_With_Decay_Pytorch community
Papers with Code archive
1 ran · 0 no recorded run
abhuse/cyclic-cosine-decay community
Papers with Code archive
1 ran · 0 no recorded run
MrtnMndt/Rethinking_CNN_Layerwise_Feature_Amounts community
Papers with Code archive
1 ran · 0 no recorded run
Jmak12/Iris1 community
Papers with Code archive
1 ran · 0 no recorded run
FunctionStatusWhere it lives
CosineAnnealingLR_with_Restart Ran
ran on a synthesized input · constructed an object; no result was computed
Gjiangtao/A-Deep-Supervised-Edge-Optimization-Algorithm-for-Salt-Body-Segmentation/config/res34_loss_naive_nodice_hp_scse_ds_border/loss/cyclic_lr.py
pointer only (licence: NONE) · get_code("4f7499d2bb391700")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Gjiangtao/A-Deep-Supervised-Edge-Optimization-Algorithm-for-Salt-Body-Segmentation 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.

CosineAnnealingWithRestartsLR Ran
ran on a synthesized input · constructed an object; no result was computed
gurucharanmk/PyTorch_CosineAnnealingWithRestartsLR/CosineAnnealingWithRestartsLR.py
code served (permissive licence) · get_code("f4a8879aa783ad15")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, CosineAnnealingWithRestartsLR: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from torch.optim import SGD

torch.manual_seed(0)

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

# Instantiate the model
model = SimpleModel()

# Create an optimizer
optimizer = SGD(model.parameters(), lr=0.1)

# Define arguments for CosineAnnealingWithRestartsLR
ARGS = {
    "optimizer": optimizer,
    "T_max": 10,
    "eta_min": 0.001,
    "last_epoch": -1,
    "T_mult": 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type CosineAnnealingWithRestartsLR — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a CosineAnnealingWithRestartsLR 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
CosineLR Ran
ran on a synthesized input · constructed an object; no result was computed
SevenZhan/Pytorch/schedulers.py
pointer only (licence: NONE) · get_code("fe5e79b70f387bec")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, SevenZhan/Pytorch when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

CyclicCosAnnealingLR Ran
ran on a synthesized input
Harshvardhan1/cyclic-learning-schedulers-pytorch/cyclicLR.py
pointer only (licence: NONE) · get_code("8a9fabc6ccb89932")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Harshvardhan1/cyclic-learning-schedulers-pytorch when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

CyclicCosAnnealingLR Ran
ran on a synthesized input
bluesky314/Cyclical_LR_Scheduler_With_Decay_Pytorch/cyclicLR.py
pointer only (licence: NONE) · get_code("43331e0456fed368")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, bluesky314/Cyclical_LR_Scheduler_With_Decay_Pytorch when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

CyclicCosineDecayLR Ran
ran on a synthesized input · constructed an object; no result was computed
abhuse/cyclic-cosine-decay/scheduler.py
code served (permissive licence) · get_code("926fab0aa32bd97a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, CyclicCosineDecayLR: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from torch.optim import SGD

torch.manual_seed(0)

# Create a simple model
model = nn.Linear(4, 3)

# Create an optimizer with the model parameters
optimizer = SGD(model.parameters(), lr=0.1)

# Define ARGS dictionary with all parameters for CyclicCosineDecayLR
ARGS = {
    'optimizer': optimizer,
    'init_decay_epochs': 10,
    'min_decay_lr': 0.001,
    'restart_interval': 5,
    'restart_interval_multiplier': 1.5,
    'restart_lr': 0.05,
    'warmup_epochs': 2,
    'warmup_start_lr': 0.001,
    'last_epoch': -1,
    'verbose': 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 CyclicCosineDecayLR — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a CyclicCosineDecayLR 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
LearningRateScheduler Ran
ran on a synthesized input · constructed an object; no result was computed
MrtnMndt/Rethinking_CNN_Layerwise_Feature_Amounts/lib/Training/learning_rate_scheduling.py
code served (permissive licence) · get_code("e2d73df88181d0fb")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, LearningRateScheduler: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import torch.optim as optim

torch.manual_seed(0)

# Create a simple optimizer with parameter groups
model = nn.Linear(4, 3)
optimizer = optim.SGD(model.parameters(), lr=0.1)

# Define all required arguments
ARGS = {
    "wr_epochs": 10,
    "dataset_size": 100,
    "batch_size": 32,
    "max_learning_rate": 0.1,
    "wr_mul": 2,
    "min_learning_rate": 1e-05,
}
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 LearningRateScheduler — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a LearningRateScheduler 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
PolyLR Ran
ran on a synthesized input · constructed an object; no result was computed
Jmak12/Iris1/optimizers/lr_scheduler.py
pointer only (licence: NONE) · get_code("899d6ddaba2af2a7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Jmak12/Iris1 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.

SGDRLR Ran
ran on a synthesized input · constructed an object; no result was computed
jae1213/CyclicalLR_SGDR/LRscheduler.py
pointer only (licence: NONE) · get_code("6e9ad0df07eed6da")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jae1213/CyclicalLR_SGDR 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.

computeIoU Ran
ran on a synthesized input
buptlwz/mabp/gres_model/evaluation/refer_evaluation.py
code served (permissive licence) · get_code("745bf09fefeb50c3")
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 numpy as np

# Create small binary segmentation masks
# pred_seg: predicted segmentation (binary)
# gd_seg: ground truth segmentation (binary)

pred_seg = np.array([
    [True, False, True, False],
    [False, True, False, True]
], dtype=bool)

gd_seg = np.array([
    [True, True, False, False],
    [False, True, True, False]
], dtype=bool)

ARGS = {
    "pred_seg": pred_seg,
    "gd_seg": gd_seg
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, 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
nested_tensor_from_tensor_list Ran
ran; our contract draft was wrong, not the code
buptlwz/mabp/gres_model/utils/misc.py
code served (permissive licence) · get_code("58cc9ff3bf75e753")
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 NestedTensor. 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.functional as F
from typing import List

torch.manual_seed(0)

# Create a list of 3D tensors with varying shapes
# Each tensor has shape (C, H, W) where C is channels, H is height, W is width
tensor1 = torch.randn(3, 4, 5)  # 3 channels, 4 height, 5 width
tensor2 = torch.randn(3, 6, 7)  # 3 channels, 6 height, 7 width
tensor3 = torch.randn(3, 5, 8)  # 3 channels, 5 height, 8 width

tensor_list = [tensor1, tensor2, tensor3]

ARGS = {
    "tensor_list": tensor_list
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type NestedTensor — a NestedTensor output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a NestedTensor 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
prepare_data Ran
ran on a synthesized input
buptlwz/mabp/gres_model/data_prepare.py
code served (permissive licence) · get_code("755b208f128fbce6")
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. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type 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)

# Construct training data
train_data = [
    {
        'image': torch.randn(3, 64, 64),
        'empty': torch.randint(0, 2, (1,)),
        'mask': torch.randint(0, 2, (64, 64)),
        'lang_tokens': torch.randint(0, 100, (5, 128)),
        'lang_masks': torch.ones(5, 128),
        'sents': 'a sample sentence'
    },
    {
        'image': torch.randn(3, 64, 64),
        'empty': torch.randint(0, 2, (1,)),
        'mask': torch.randint(0, 2, (64, 64)),
        'lang_tokens': torch.randint(0, 100, (5, 128)),
        'lang_masks': torch.ones(5, 128),
        'sents': 'another sentence'
    }
]

# Construct inference data (single batch)
infer_data = [
    {
        'image': torch.randn(3, 64, 64),
        'empty': torch.randint(0, 2, (1,)),
        'mask': torch.randint(0, 2, (64, 64)),
        'lang_tokens': [torch.randint(0, 100, (3, 128)), torch.randint(0, 100, (4, 128))],
        'lang_masks': [torch.ones(3, 128), torch.ones(4, 128)],
        'sents': 'inference sentence'
    }
]

ARGS = {
    'data': train_data,
    'train': 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 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
refer_bce_loss Ran
ran on a synthesized input
buptlwz/mabp/gres_model/modeling/criterion-so.py
code served (permissive licence) · get_code("67dae4afd1febb5e")
How it ran

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

import torch
import torch.nn.functional as F

torch.manual_seed(0)

# Create small tensors for inputs and targets
# inputs: logits tensor (batch_size=2, num_classes=4)
inputs = torch.randn(2, 4)

# targets: binary labels tensor (batch_size=2, num_classes=4)
targets = torch.randint(0, 2, (2, 4)).float()

# weight parameter is None according to function signature
weight = None

ARGS = {
    "inputs": inputs,
    "targets": targets,
    "weight": weight
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
refer_ce_loss Ran
ran on a synthesized input
buptlwz/mabp/gres_model/modeling/criterion-so.py
code served (permissive licence) · get_code("7b44eda6b63c9a3a")
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 torch.nn.functional as F

torch.manual_seed(0)

# refer_ce_loss expects:
# - inputs: torch.Tensor (logits, shape [batch_size, num_classes])
# - targets: torch.Tensor (class indices, shape [batch_size])
# - weight: torch.Tensor (per-class weights, shape [num_classes])

batch_size = 2
num_classes = 4

inputs = torch.randn(batch_size, num_classes)
targets = torch.tensor([0, 2])
weight = torch.ones(num_classes)

ARGS = {
    "inputs": inputs,
    "targets": targets,
    "weight": weight,
}
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
window_partition Ran
ran on a synthesized input
buptlwz/mabp/gres_model/modeling/criterion-so.py
code served (permissive licence) · get_code("2f4f1157ededdc6b")
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 5, dtype family float, shape (2, 4, 4, 4, 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 as nn

torch.manual_seed(0)

# Create a small input tensor with shape (B, H, W, C)
# B=2, H=8, W=8, C=4 (window_size will be 4, so H and W must be divisible by it)
x = torch.randn(2, 8, 8, 4)

# window_size must divide both H and W evenly
window_size = 4

ARGS = {
    "x": x,
    "window_size": window_size,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 5, dtype family float, shape (2, 4, 4, 4, 4) — shape (2, 4, 4, 4, 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
window_reverse Ran
ran; our contract draft was wrong, not the code
buptlwz/mabp/gres_model/modeling/backbone/swin.py
code served (permissive licence) · get_code("61bf152e6a42a184")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 4, dtype family float, shape (2, 8, 12, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Parameters for window_reverse
# windows: (num_windows*B, window_size, window_size, C)
# The function expects windows to be partitioned from an image of size (B, H, W, C)
# where the image is divided into non-overlapping windows of size window_size x window_size

B = 2  # batch size
H = 8  # image height
W = 12  # image width
window_size = 4  # window size
C = 3  # number of channels

# Calculate number of windows
num_windows_h = H // window_size  # 2
num_windows_w = W // window_size  # 3
num_windows = num_windows_h * num_windows_w  # 6

# windows shape: (num_windows*B, window_size, window_size, C)
# = (6*2, 4, 4, 3) = (12, 4, 4, 3)
windows = torch.randn(num_windows * B, window_size, window_size, C)

ARGS = {
    "windows": windows,
    "window_size": window_size,
    "H": H,
    "W": W,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 8, 12, 3) — shape (2, 8, 12, 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
CustomLR Not yet run
unverified (harvested, no recorded run)
jae1213/CyclicalLR_SGDR/LRscheduler.py
pointer only (licence: NONE) · get_code("803f499019d43ed0")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jae1213/CyclicalLR_SGDR 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.

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 later) and 14 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
window_reverse 227 other papers
227 later · 34 only as an assembly brick
1711.10925 (2017-11)
Deep Image Prior
2003.13678 (2020-03)
Designing Network Design Spaces
2007.09384 (2020-07)
Multi-Scale Positive Sample Refinement for Few-Shot Object Detection
and 224 more not named here
nested_tensor_from_tensor_list 55 other papers
55 later · 17 only as an assembly brick
2005.12872 (2020-05)
End-to-End Object Detection with Transformers
2101.01909 (2021-01)
Line Segment Detection Using Transformers without Edges
2103.04503 (2021-03)
End-to-End Human Object Interaction Detection with HOI Transformer
and 52 more not named here
computeIoU 1 other paper
1 later
2306.00968 (2023-06)
GRES: Generalized Referring Expression Segmentation

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

Restart techniques are common in gradient-free optimization to deal with multimodal functions. Partial warm restarts are also gaining popularity in gradientbased optimization to improve the rate of convergence in accelerated gradient schemes to deal with ill-conditioned functions. In this paper, we propose a simple warm restart technique for stochastic gradient descent to improve its anytime performance when training deep neural networks. We empirically study its performance on the CIFAR-10 and CIFAR-100 datasets, where we demonstrate new state-of-the-art results at 3.14% and 16.21%, respectively. We also demonstrate its advantages on a dataset of EEG recordings and on a downsampled version of the ImageNet dataset. Our source code is available at https://github.com/loshchil/SGDR

For agents

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

get_harvested_code_for_paper("1608.03983")
get_code_for_paper("1608.03983")
have("1608.03983")

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

curl https://syntology.ai/api/ran/1608.03983.json

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

[![Syntology run record](https://syntology.ai/api/ran/1608.03983.svg)](https://syntology.ai/paper/1608.03983)

Connect an agent — have() is free.