SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1802.10026 · 2018

Loss Surfaces, Mode Connectivity, and Fast Ensembling of DNNs

arXiv · PDF · Open in the Atlas

Code that ran

8 of 14 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; 2 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 3 failures of our instrument; 6 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/1802.10026.json · badge: /api/ran/1802.10026.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
tjwhitaker/prune-and-tune-ensembles community (classifier label)
classifier basis not recorded on the link
5 ran · 5 no recorded run
xuyxu/Ensemble-Pytorch community (classifier label)
classifier basis not recorded on the link
3 ran · 1 no recorded run
FunctionStatusWhere it lives
adjust_learning_rate Ran
ran, honoured the contract we drafted
tjwhitaker/prune-and-tune-ensembles/src/utils.py
code served (permissive licence) · get_code("14ebf34e2f003912")
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 float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

class DummyModel(nn.Module):
    def __init__(self):
        super(DummyModel, self).__init__()
        self.fc = nn.Linear(4, 3)

model = DummyModel()
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)
lr = 0.001

ARGS = {
    'optimizer': optimizer,
    'lr': lr
}
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 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
conv1x1 Ran
ran; our contract draft was wrong, not the code
tjwhitaker/prune-and-tune-ensembles/src/models/resnet.py
code served (permissive licence) · get_code("d9def42110729a85")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Conv2d. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

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

ARGS = {
    "in_planes": 3,
    "out_planes": 64,
    "stride": 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
conv3x3 Ran
ran; our contract draft was wrong, not the code
tjwhitaker/prune-and-tune-ensembles/src/models/resnet.py
code served (permissive licence) · get_code("160bb14bd76201b4")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Conv2d. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "in_planes": 3,
    "out_planes": 64,
    "stride": 1,
    "groups": 1,
    "dilation": 1,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
densenet169 Ran
ran; our contract draft was wrong, not the code
tjwhitaker/prune-and-tune-ensembles/src/models/densenet.py
code served (permissive licence) · get_code("9ee8bc3f789e311e")
How it ran

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

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'pretrained': False,
    'progress': True,
    'device': 'cpu',
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DenseNet — a DenseNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DenseNet output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
learning_rate_schedule Ran
ran, honoured the contract we drafted
tjwhitaker/prune-and-tune-ensembles/src/utils.py
code served (permissive licence) · get_code("1bdf337d2589e11b")
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 torch

torch.manual_seed(0)

# The function takes three scalar parameters: base_lr, epoch, and total_epochs
# It performs arithmetic operations on them to compute a learning rate schedule factor
# All parameters are numeric values

base_lr = 0.1
epoch = 25
total_epochs = 100

ARGS = {
    "base_lr": base_lr,
    "epoch": epoch,
    "total_epochs": total_epochs
}
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 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
set_scheduler Ran
ran on a synthesized input
xuyxu/Ensemble-Pytorch/torchensemble/utils/set_module.py
code served (permissive licence) · get_code("bb98a99452b24eae")
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 StepLR. 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 model
model = nn.Linear(4, 3)

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

# Define scheduler_name
scheduler_name = 'StepLR'

# Define kwargs for StepLR scheduler
kwargs = {
    'step_size': 10,
    'gamma': 0.1
}

# Construct ARGS dict
ARGS = {
    'optimizer': optimizer,
    'scheduler_name': scheduler_name,
    **kwargs
}
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 StepLR — a StepLR output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a StepLR 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
torchensemble_model_doc Ran
ran on a synthesized input
xuyxu/Ensemble-Pytorch/torchensemble/_base.py
code served (permissive licence) · get_code("89aa135e73e6bce8")
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

# Create a minimal const module with the required documentation strings
class const:
    __model_doc = ["Model documentation for base model."]
    __seq_model_doc = ["Sequential model documentation."]
    __tree_ensemble_doc = ["Tree ensemble model documentation."]
    __fit_doc = ["Fit method documentation."]
    __predict_doc = ["Predict method documentation."]
    __set_optimizer_doc = ["Set optimizer documentation."]
    __set_scheduler_doc = ["Set scheduler documentation."]
    __set_criterion_doc = ["Set criterion documentation."]
    __classification_forward_doc = ["Classification forward documentation."]
    __classification_evaluate_doc = ["Classification evaluate documentation."]
    __regression_forward_doc = ["Regression forward documentation."]
    __regression_evaluate_doc = ["Regression evaluate documentation."]

ARGS = {
    'header': 'Ensemble Model Documentation',
    'item': '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 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
update_lr Ran
ran on a synthesized input
xuyxu/Ensemble-Pytorch/torchensemble/utils/set_module.py
code served (permissive licence) · get_code("e85ee74b940ce638")
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 SGD. 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 model
model = nn.Linear(4, 3)

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

# Define ARGS with valid arguments for update_lr
ARGS = {
    'optimizer': optimizer,
    'lr': 0.01
}
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 SGD — a SGD output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SGD 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
densenet121 Not yet run
unverified (harvested, no recorded run)
tjwhitaker/prune-and-tune-ensembles/src/models/densenet.py
code served (permissive licence) · get_code("d2ef37da818b9f51")
How it ran

no invocation recorded.

densenet161 Not yet run
unverified (harvested, no recorded run)
tjwhitaker/prune-and-tune-ensembles/src/models/densenet.py
code served (permissive licence) · get_code("2c599b41c3a242e3")
How it ran

no invocation recorded.

get_corrupted_loaders Not yet run
unverified (harvested, no recorded run)
tjwhitaker/prune-and-tune-ensembles/src/data.py
code served (permissive licence) · get_code("8e5a659674eb5d9c")
How it ran

no invocation recorded.

get_loaders Not yet run
unverified (harvested, no recorded run)
tjwhitaker/prune-and-tune-ensembles/src/data.py
code served (permissive licence) · get_code("4b373666065863fd")
How it ran

no invocation recorded.

resnet18 Not yet run
unverified (harvested, no recorded run)
tjwhitaker/prune-and-tune-ensembles/src/models/resnet.py
code served (permissive licence) · get_code("2615428ce71ead9f")
How it ran

no invocation recorded.

set_optimizer Not yet run
unverified (harvested, no recorded run)
xuyxu/Ensemble-Pytorch/torchensemble/utils/set_module.py
code served (permissive licence) · get_code("c5a3a3cd326f2289")
How it ran

no invocation recorded.

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 8 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 8 later) and 6 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
conv1x1 431 other papers
13 dated earlier, 417 later, 1 undated · 38 only as an assembly brick
1409.1556 (2014-09)
Very Deep Convolutional Networks for Large-Scale Image Recognition
1506.01186 (2015-06)
Cyclical Learning Rates for Training Neural Networks
1507.05717 (2015-07)
An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition
and 428 more not named here
conv3x3 312 other papers
5 dated earlier, 306 later, 1 undated · 31 only as an assembly brick
1409.1556 (2014-09)
Very Deep Convolutional Networks for Large-Scale Image Recognition
1507.05717 (2015-07)
An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition
and 309 more not named here
resnet18 6 other papers
6 later
2206.07290 (2022-06)
Differentiable Top-k Classification Learning
2301.11578 (2023-01)
Learning to Unlearn: Instance-wise Unlearning for Pre-trained Classifiers
2403.01101 (2024-03)
Feature Alignment: Rethinking Efficient Active Learning via Proxy in the Context of Pre-trained Models
and 3 more not named here
adjust_learning_rate 4 other papers
4 later
1803.05407 (2018-03)
Averaging Weights Leads to Wider Optima and Better Generalization
1812.10366 (2018-12)
A Poisson-Gaussian Denoising Dataset with Real Fluorescence Microscopy Images
2109.07839 (2021-09)
Self-supervised Contrastive Learning for EEG-based Sleep Staging
and 1 more not named here
densenet121 3 other papers
3 later
2204.03714 (2022-04)
ICLR 2022 PAIR 2 Struct Workshop USING MULTIPLE SELF-SUPERVISED TASKS IM-PROVES MODEL ROBUSTNESS
2406.15093 (2024-06)
ECLIPSE: Expunging Clean-label Indiscriminate Poisons via Sparse Diffusion Purification
2505.05522 (2025-05)
Continuous Thought Machines
densenet161 3 other papers
3 later
2204.03714 (2022-04)
ICLR 2022 PAIR 2 Struct Workshop USING MULTIPLE SELF-SUPERVISED TASKS IM-PROVES MODEL ROBUSTNESS
2406.15093 (2024-06)
ECLIPSE: Expunging Clean-label Indiscriminate Poisons via Sparse Diffusion Purification
2505.05522 (2025-05)
Continuous Thought Machines
densenet169 3 other papers
3 later
2204.03714 (2022-04)
ICLR 2022 PAIR 2 Struct Workshop USING MULTIPLE SELF-SUPERVISED TASKS IM-PROVES MODEL ROBUSTNESS
2406.15093 (2024-06)
ECLIPSE: Expunging Clean-label Indiscriminate Poisons via Sparse Diffusion Purification
2505.05522 (2025-05)
Continuous Thought Machines
learning_rate_schedule 1 other paper
1 later
2107.11228 (2021-07)
Taxonomizing local versus global structure in neural network loss landscapes

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

The loss functions of deep neural networks are complex and their geometric properties are not well understood. We show that the optima of these complex loss functions are in fact connected by simple curves over which training and test accuracy are nearly constant. We introduce a training procedure to discover these high-accuracy pathways between modes. Inspired by this new geometric insight, we also propose a new ensembling method entitled Fast Geometric Ensembling (FGE). Using FGE we can train high-performing ensembles in the time required to train a single model. We achieve improved performance compared to the recent state-of-the-art Snapshot Ensembles, on CIFAR-10, CIFAR-100, and ImageNet.

For agents

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

get_harvested_code_for_paper("1802.10026")
get_code_for_paper("1802.10026")
have("1802.10026")

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

curl https://syntology.ai/api/ran/1802.10026.json

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

[![Syntology run record](https://syntology.ai/api/ran/1802.10026.svg)](https://syntology.ai/paper/1802.10026)

Connect an agent — have() is free.