SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2202.03990 · ICML · 2022

Equivariance versus Augmentation for Spherical Images

Jan Gerken, Oscar Carlsson, Hampus Linander, Fredrik Ohlsson, Christoffer Petersson, Daniel Persson

arXiv · PDF · Open in the Atlas

Code that ran

5 of 9 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 5 have no contract verdict; 0 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/2202.03990.json · badge: /api/ran/2202.03990.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
janegerken/sem_seg_s2cnn canonical (classifier label)
classifier basis not recorded on the link
5 ran · 4 no recorded run
FunctionStatusWhere it lives
s2_equatorial_grid Ran
ran on a synthesized input
janegerken/sem_seg_s2cnn/s2cnn/s2_grid.py
code served (permissive licence) · get_code("093cf675bf110cfd")
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 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

ARGS = {
    'max_beta': 0.5,
    'n_alpha': 32,
    'n_beta': 1
}
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 2, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
s2_soft_grid Ran
ran on a synthesized input
janegerken/sem_seg_s2cnn/s2cnn/s2_grid.py
code served (permissive licence) · get_code("7135ee0ecb855508")
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 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

def s2_soft_grid(b):
    beta = (np.arange(2 * b) + 0.5) / (2 * b) * np.pi
    alpha = np.linspace(start=0, stop=2 * np.pi, num=2 * b, endpoint=False)
    B, A = np.meshgrid(beta, alpha, indexing='ij')
    B = B.flatten()
    A = A.flatten()
    grid = np.stack((B, A), axis=1)
    return tuple((tuple(ba) for ba in grid))

ARGS = {
    "b": 3
}
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 2, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
so3_equatorial_grid Ran
ran on a synthesized input
janegerken/sem_seg_s2cnn/s2cnn/so3_grid.py
code served (permissive licence) · get_code("f8b54487de19b999")
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 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

ARGS = {
    'max_beta': 0,
    'max_gamma': np.pi / 8,
    'n_alpha': 32,
    'n_beta': 1,
    'n_gamma': 2,
}
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 2, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
so3_soft_grid Ran
ran on a synthesized input
janegerken/sem_seg_s2cnn/s2cnn/so3_grid.py
code served (permissive licence) · get_code("55908293a06be7ea")
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 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

def so3_soft_grid(b):
    beta = (np.arange(2 * b) + 0.5) / (2 * b) * np.pi
    alpha = gamma = np.linspace(start=0, stop=2 * np.pi, num=2 * b, endpoint=False)
    B, A, C = np.meshgrid(beta, alpha, gamma, indexing='ij')
    B = B.flatten()
    A = A.flatten()
    C = C.flatten()
    grid = np.stack((B, A, C), axis=1)
    return tuple((tuple(bac) for bac in grid))

ARGS = {
    'b': 2
}
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 2, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
so3_sum_n Ran
ran on a synthesized input
janegerken/sem_seg_s2cnn/s2cnn/so3_sum_n.py
code served (permissive licence) · get_code("b313e9a26d0f5717")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 4, dtype family float, shape (4, 2, 4, 2) as a later run recorded it. 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, 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 math

torch.manual_seed(0)

def complex_sum(x, dim):
    """Stand-in for s2cnn.utils.complex.complex_sum"""
    return torch.sum(x, dim=dim)

# Monkey-patch the complex_sum into the s2cnn module structure
import sys
from types import ModuleType

s2cnn = ModuleType('s2cnn')
s2cnn_utils = ModuleType('s2cnn.utils')
s2cnn_utils_complex = ModuleType('s2cnn.utils.complex')
s2cnn_utils_complex.complex_sum = complex_sum

s2cnn_utils.complex = s2cnn_utils_complex
s2cnn.utils = s2cnn_utils
sys.modules['s2cnn'] = s2cnn
sys.modules['s2cnn.utils'] = s2cnn_utils
sys.modules['s2cnn.utils.complex'] = s2cnn_utils_complex

# Now construct valid arguments
# The function expects x with shape [l * m * n, batch, feature_in, complex]
# where nspec = nl * (4 * nl^2 - 1) / 3
# Let's use nl = 2:
# nspec = 2 * (4 * 4 - 1) / 3 = 2 * 15 / 3 = 10

nl = 2
nspec = nl * (4 * nl ** 2 - 1) // 3  # = 10
nbatch = 2
nfeature_in = 4
complex_dim = 2

x = torch.randn(nspec, nbatch, nfeature_in, complex_dim)

ARGS = {
    'x': x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
s2cnn, 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 (4, 2, 4, 2) — shape (4, 2, 4, 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
s2_mm Not yet run
unverified (harvested, no recorded run)
janegerken/sem_seg_s2cnn/s2cnn/s2_mm.py
code served (permissive licence) · get_code("1015e15faa793a33")
How it ran

no invocation recorded.

s2_near_identity_grid Not yet run
unverified (harvested, no recorded run)
janegerken/sem_seg_s2cnn/s2cnn/s2_grid.py
code served (permissive licence) · get_code("165f749be67e7f85")
How it ran

no invocation recorded.

so3_mm Not yet run
unverified (harvested, no recorded run)
janegerken/sem_seg_s2cnn/s2cnn/so3_mm.py
code served (permissive licence) · get_code("104eed4c50b2b84a")
How it ran

no invocation recorded.

so3_near_identity_grid Not yet run
unverified (harvested, no recorded run)
janegerken/sem_seg_s2cnn/s2cnn/so3_grid.py
code served (permissive licence) · get_code("a54c4f20182df475")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 9 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.

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

We analyze the role of rotational equivariance in convolutional neural networks (CNNs) applied to spherical images. We compare the performance of the group equivariant networks known as S2CNNs and standard non-equivariant CNNs trained with an increasing amount of data augmentation. The chosen architectures can be considered baseline references for the respective design paradigms. Our models are trained and evaluated on single or multiple items from the MNIST or FashionMNIST dataset projected onto the sphere. For the task of image classification, which is inherently rotationally invariant, we find that by considerably increasing the amount of data augmentation and the size of the networks, it is possible for the standard CNNs to reach at least the same performance as the equivariant network. In contrast, for the inherently equivariant task of semantic segmentation, the non-equivariant networks are consistently outperformed by the equivariant networks with significantly fewer parameters. We also analyze and compare the inference latency and training times of the different networks, enabling detailed tradeoff considerations between equivariant architectures and data augmentation for practical problems. The equivariant spherical networks used in the experiments are available at https://github. com/JanEGerken/sem_seg_s2cnn.

For agents

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

get_harvested_code_for_paper("2202.03990")
get_code_for_paper("2202.03990")
have("2202.03990")

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

curl https://syntology.ai/api/ran/2202.03990.json

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

[![Syntology run record](https://syntology.ai/api/ran/2202.03990.svg)](https://syntology.ai/paper/2202.03990)

Connect an agent — have() is free.