SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2006.10503 · 2020

SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networks

arXiv · PDF · Open in the Atlas

Code that ran

22 of 28 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 21 have no contract verdict; 1 failure 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/2006.10503.json · badge: /api/ran/2006.10503.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
rosettacommons/rosettafold community (classifier label)
classifier basis not recorded on the link
14 ran · 3 no recorded run
trachote/predict_elastic_tensor community (classifier label)
classifier basis not recorded on the link
7 ran · 3 no recorded run
lucidrains/se3-transformer-pytorch reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
count_parameters Ran
ran on a synthesized input
trachote/predict_elastic_tensor/se3t/utils/utils_logging.py
code served (permissive licence) · get_code("29e2228c42add029")
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 int64, ndim 0, dtype family int, 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
import numpy as np

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

model = SimpleModel()

ARGS = {
    "model": model
}
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 int64, ndim 0, dtype family int, 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
create_custom_forward Ran
ran on a synthesized input
rosettacommons/rosettafold/network/Transformer.py
code served (permissive licence) · get_code("c68c3daa8125c3e8")
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

torch.manual_seed(0)

class SimpleModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x, y=None, scale=1.0):
        result = self.linear(x)
        if y is not None:
            result = result + y
        return result * scale

module = SimpleModule()

ARGS = {
    "module": module,
    "scale": 2.0,
    "offset": 0.5
}
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
fiber2tensor Ran
ran on a synthesized input
trachote/predict_elastic_tensor/se3t/equivariant_attention/fibers.py
code served (permissive licence) · get_code("217d7fa95a8654fb")
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 (2, 4, 54, 1) 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 structure object that has a 'degrees' attribute
class Structure:
    def __init__(self, degrees):
        self.degrees = degrees

# Create sample fiber data dictionary
# The function expects F to be a dict with keys like '0', '1', '2', etc.
# Each value should be a tensor with shape (*batch_shape, ..., fiber_dim1, fiber_dim2)
F = {
    '0': torch.randn(2, 4, 3, 5),  # batch=2, feature=4, fiber_dim1=3, fiber_dim2=5
    '1': torch.randn(2, 4, 3, 7),  # batch=2, feature=4, fiber_dim1=3, fiber_dim2=7
    '2': torch.randn(2, 4, 3, 6),  # batch=2, feature=4, fiber_dim1=3, fiber_dim2=6
}

structure = Structure(degrees=['0', '1', '2'])

ARGS = {
    'F': F,
    'structure': structure,
    'squeeze': 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 Tensor, ndim 4, dtype family float, shape (2, 4, 54, 1) — shape (2, 4, 54, 1) 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_ang Ran
ran on a synthesized input
rosettacommons/rosettafold/network/kinematics.py
code served (permissive licence) · get_code("9f0e16c4abfe096f")
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

torch.manual_seed(0)

# Create small tensors of shape [batch, nres, 3]
# batch=2, nres=4, 3 coordinates (x, y, z)
batch_size = 2
nres = 4
coord_dim = 3

a = torch.randn(batch_size, nres, coord_dim)
b = torch.randn(batch_size, nres, coord_dim)
c = torch.randn(batch_size, nres, coord_dim)

ARGS = {
    "a": a,
    "b": b,
    "c": c,
}
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
get_bonded_neigh Ran
ran on a synthesized input
rosettacommons/rosettafold/network/Attention_module_w_str.py
code served (permissive licence) · get_code("4eb618ae14addf59")
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 (2, 4, 4, 1) 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)

# idx should be a tensor of shape (B, L) containing residue indices
# B = batch size, L = sequence length
# For this function, the values represent residue indices that will be subtracted
idx = torch.tensor([
    [0, 1, 2, 3],
    [0, 1, 2, 3]
], dtype=torch.long)

ARGS = {
    "idx": idx
}
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 4, dtype family float, shape (2, 4, 4, 1) — shape (2, 4, 4, 1) 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_dih Ran
ran on a synthesized input
rosettacommons/rosettafold/network/kinematics.py
code served (permissive licence) · get_code("a71c2f40de637572")
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

torch.manual_seed(0)

# Create tensors of shape [batch, nres, 3]
# batch=2, nres=4, 3D coordinates
batch_size = 2
nres = 4
coord_dim = 3

a = torch.randn(batch_size, nres, coord_dim)
b = torch.randn(batch_size, nres, coord_dim)
c = torch.randn(batch_size, nres, coord_dim)
d = torch.randn(batch_size, nres, coord_dim)

ARGS = {
    'a': a,
    'b': b,
    'c': c,
    'd': d
}
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
get_entry_by_name Ran
ran on a synthesized input
rosettacommons/rosettafold/network/ffindex.py
code served (permissive licence) · get_code("dac71ff1e8526ef4")
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 Entry. 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

class Entry:
    def __init__(self, name):
        self.name = name

# Create a list of entries to serve as the index
entry1 = Entry("apple")
entry2 = Entry("banana")
entry3 = Entry("cherry")
index = [entry1, entry2, entry3]

# Define arguments for get_entry_by_name
ARGS = {
    "name": "banana",
    "index": index
}
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 Entry — a Entry output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Entry 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
get_epsilon Ran
ran on a synthesized input
trachote/predict_elastic_tensor/utils/etc.py
code served (permissive licence) · get_code("767887f5d7462380")
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 4, dtype family float, shape (6, 6, 3, 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 numpy as np

# The function get_epsilon takes a single parameter: strain
# which should be a numeric value (int or float)
# The function performs numpy operations on it to compute strain-related matrices

strain_value = 2.5

ARGS = {
    "strain": strain_value
}
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 4, dtype family float, shape (6, 6, 3, 3) — shape (6, 6, 3, 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
get_pair_dist Ran
ran on a synthesized input
rosettacommons/rosettafold/network/kinematics.py
code served (permissive licence) · get_code("0ae837110bf18cc4")
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 3, dtype family float, shape (2, 4, 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

torch.manual_seed(0)

# Create two sets of point clouds
# Shape: [batch, nres, 3] where batch=2, nres=4, 3 is for Cartesian coordinates
a = torch.randn(2, 4, 3)
b = torch.randn(2, 5, 3)

ARGS = {
    'a': a,
    'b': b,
}
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 3, dtype family float, shape (2, 4, 5) — shape (2, 4, 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
get_seqsep Ran
ran on a synthesized input
rosettacommons/rosettafold/network/InitStrGenerator.py
code served (permissive licence) · get_code("6af62bbb3aa5d036")
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 (2, 4, 4, 1) 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)

# idx: residue indices of given sequence (B, L)
# B = batch size = 2
# L = sequence length = 4
idx = torch.arange(4, dtype=torch.long).unsqueeze(0).repeat(2, 1)

ARGS = {
    "idx": idx,
}
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 4, dtype family float, shape (2, 4, 4, 1) — shape (2, 4, 4, 1) 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_tensor_device_and_dtype Ran
ran; our contract draft was wrong, not the code
lucidrains/se3-transformer-pytorch/se3_transformer_pytorch/se3_transformer_pytorch.py
code served (permissive licence) · get_code("00b9e7bd06dbbb81")
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 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)

# The function expects a dict-like object where:
# - It calls .items() on it
# - The values are tensors (it accesses .device and .dtype)

features = {
    "input_ids": torch.tensor([[1, 2, 3, 4], [5, 6, 7, 8]], dtype=torch.float32),
    "attention_mask": torch.tensor([[1, 1, 1, 1], [1, 1, 0, 0]], dtype=torch.long),
}

ARGS = {
    "features": features,
}
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 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
lrQres2CAdev Ran
ran on a synthesized input
rosettacommons/rosettafold/DAN-msa/pick_final_models.div.py
code served (permissive licence) · get_code("17b99f2bea9c731b")
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 1, 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

# Set random seeds for reproducibility
np.random.seed(0)

# Number of residues
nres = 10

# Create estogram: shape (nres, nres, 12)
# This represents distance distribution predictions with 12 bins
estogram = np.random.rand(nres, nres, 12)

# Create mask: shape (nres, nres)
# Represents contact probability between residues
mask = np.random.rand(nres, nres)

# Set some mask values to be significant (> 0.1) to create contacts
for i in range(nres):
    for j in range(nres):
        if abs(i - j) >= 13:  # Only set contacts outside minseqsep
            mask[i][j] = np.random.rand() * 0.8 + 0.2  # Values between 0.2 and 1.0

# minseqsep parameter
minseqsep = 13

ARGS = {
    'estogram': estogram,
    'mask': mask,
    'minseqsep': minseqsep
}
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 1, 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
make_graph Ran
ran on a synthesized input
rosettacommons/rosettafold/network/InitStrGenerator.py
code served (permissive licence) · get_code("6239eb2a3d6fa13e")
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 Data. 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
from torch_geometric.data import Data

torch.manual_seed(0)

# Parameters based on function signature: make_graph(node, idx, emb)
# The function expects:
# - node: reshaped to (B*L, -1), so needs shape (B, L, D) where D is node feature dim
# - idx: used in idx[:, None, :] - idx[:, :, None], so shape (B, L)
# - emb: has .device, .shape[:2] gives (B, L), indexed as emb[b, i, j], so shape (B, L, E)

B = 2  # batch size
L = 3  # sequence length
D = 4  # node feature dimension
E = 5  # edge attribute dimension

# Create node tensor: shape (B, L, D)
node = torch.randn(B, L, D)

# Create idx tensor: shape (B, L) with integer indices
idx = torch.arange(L).unsqueeze(0).expand(B, -1).float()

# Create emb tensor: shape (B, L, E)
emb = torch.randn(B, L, E)

ARGS = {
    "node": node,
    "idx": idx,
    "emb": emb,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torch_geometric — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Data — a Data 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
make_logdir Ran
ran on a synthesized input
trachote/predict_elastic_tensor/se3t/utils/utils_logging.py
code served (permissive licence) · get_code("133a4d7dcf9d6536")
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 os
import tempfile
import shutil

# Create a temporary directory for testing
test_checkpoint_dir = tempfile.mkdtemp()

# Define ARGS with all parameters of make_logdir
ARGS = {
    'checkpoint_dir': test_checkpoint_dir,
    'run_name': 'test_run_2024_01_15_10.30.45'
}
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 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
parse_pdb Ran
ran on a synthesized input
rosettacommons/rosettafold/network/parsers.py
code served (permissive licence) · get_code("644ba2df038852fc")
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 numpy as np
import tempfile
import os

# Create a minimal valid PDB file
pdb_content = """ATOM      1  N   ALA A   1      10.000  20.000  30.000  1.00  0.00           N
ATOM      2  CA  ALA A   1      11.000  21.000  31.000  1.00  0.00           C
ATOM      3  C   ALA A   1      12.000  22.000  32.000  1.00  0.00           C
ATOM      4  N   GLY A   2      13.000  23.000  33.000  1.00  0.00           N
ATOM      5  CA  GLY A   2      14.000  24.000  34.000  1.00  0.00           C
ATOM      6  C   GLY A   2      15.000  25.000  35.000  1.00  0.00           C
"""

# Write to a temporary file
with tempfile.NamedTemporaryFile(mode='w', suffix='.pdb', delete=False) as f:
    f.write(pdb_content)
    temp_pdb_file = f.name

try:
    ARGS = {
        'filename': temp_pdb_file
    }
finally:
    # Clean up will happen after execution
    pass
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 — 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
rbf Ran
ran on a synthesized input
rosettacommons/rosettafold/network/Attention_module_w_str.py
code served (permissive licence) · get_code("7db5bc710b7682dc")
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 3, dtype family float, shape (2, 4, 36) 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)

# D should be a tensor of pairwise distances
# The function expects D to have a device attribute and support unsqueeze
# Looking at the usage: D_expand = torch.unsqueeze(D, -1)
# This suggests D is a 2D tensor of distances (e.g., batch_size x num_pairs)
D = torch.randn(2, 4)

ARGS = {
    "D": D
}
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 3, dtype family float, shape (2, 4, 36) — shape (2, 4, 36) 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
read_data Ran
ran on a synthesized input
rosettacommons/rosettafold/network/ffindex.py
code served (permissive licence) · get_code("897dff3bc4c4b614")
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 mmap. 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 mmap
import tempfile
import os

# Create a temporary file with some binary data
temp_fd, temp_path = tempfile.mkstemp()
try:
    # Write some test data to the temporary file
    os.write(temp_fd, b"test binary data for mmap")
    os.close(temp_fd)
    
    # Now construct the arguments
    ARGS = {
        "ffdata_filename": temp_path
    }
finally:
    # Clean up after defining ARGS so the file exists when the function is called
    pass
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 mmap — a mmap output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a mmap output the recorded ndim 1 and dtype family int 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
read_index Ran
ran on a synthesized input
rosettacommons/rosettafold/network/ffindex.py
code served (permissive licence) · get_code("205feae362fba7d4")
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 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 tempfile
import os
from collections import namedtuple

FFindexEntry = namedtuple('FFindexEntry', 'name, offset, length')

def read_index(ffindex_filename):
    entries = []
    fh = open(ffindex_filename)
    for line in fh:
        tokens = line.split('\t')
        entries.append(FFindexEntry(tokens[0], int(tokens[1]), int(tokens[2])))
    fh.close()
    return entries

# Create a temporary ffindex file with valid content
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.ffindex')
temp_file.write("protein1\t0\t100\n")
temp_file.write("protein2\t100\t150\n")
temp_file.write("protein3\t250\t75\n")
temp_file_path = temp_file.name
temp_file.close()

ARGS = {
    'ffindex_filename': temp_file_path
}
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 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
smooth Ran
ran on a synthesized input
rosettacommons/rosettafold/DAN-msa/pick_final_models.div.py
code served (permissive licence) · get_code("0ff7824f74d12394")
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 1, dtype family float, shape (10,) 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

# Create a simple numpy array to smooth
x = np.array([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0])

ARGS = {
    'x': x,
    'window_len': 13,
    'window': 'hanning'
}
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 1, dtype family float, shape (10,) — shape (10,) 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
split_dataset Ran
ran on a synthesized input
trachote/predict_elastic_tensor/utils/datamodule.py
code served (permissive licence) · get_code("d7a94dcfe704c989")
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 numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split

# Create a minimal valid dataframe that satisfies the function requirements
np.random.seed(0)

# Create sample data with the required columns
n_samples = 10
df = pd.DataFrame({
    'elasticity.K_Voigt': np.random.uniform(0, 1500, n_samples),
    'some_edge_style_connected_graph': np.ones(n_samples, dtype=bool),
    'another_edge_style_connected_graph': np.ones(n_samples, dtype=bool),
})

ARGS = {
    'df': df,
    'val_ratio': 0.3,
    'edge_style': 'some_edge_style',
    'seed': 0,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas, sklearn — 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
strain_vector Ran
ran on a synthesized input
trachote/predict_elastic_tensor/utils/etc.py
code served (permissive licence) · get_code("6e03953b326a94ba")
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 3, dtype family float, shape (6, 6, 6) 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

ARGS = {
    "strain": 0.5
}
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 3, dtype family float, shape (6, 6, 6) — shape (6, 6, 6) 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
to_np Ran
ran on a synthesized input
trachote/predict_elastic_tensor/se3t/utils/utils_data.py
code served (permissive licence) · get_code("a95efb537cb231f0")
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, 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 numpy as np

torch.manual_seed(0)

# to_np expects a tensor x with .cpu(), .detach(), and .numpy() methods
# Build a small tensor on CPU
x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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 ndarray, 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
calc_lddt_dist Not yet run
unverified (harvested, no recorded run)
rosettacommons/rosettafold/DAN-msa/pick_final_models.div.py
code served (permissive licence) · get_code("cb4619ef753b758f")
How it ran

no invocation recorded.

get_fiber_dict Not yet run
unverified (harvested, no recorded run)
trachote/predict_elastic_tensor/se3t/equivariant_attention/fibers.py
code served (permissive licence) · get_code("4f20b0ab718934de")
How it ran

no invocation recorded.

get_fiber_tensor Not yet run
unverified (harvested, no recorded run)
trachote/predict_elastic_tensor/se3t/equivariant_attention/fibers.py
code served (permissive licence) · get_code("607f975107d53b59")
How it ran

no invocation recorded.

parse_a3m Not yet run
unverified (harvested, no recorded run)
rosettacommons/rosettafold/network/parsers.py
code served (permissive licence) · get_code("f1184128e675fe8d")
How it ran

no invocation recorded.

parse_hhr Not yet run
unverified (harvested, no recorded run)
rosettacommons/rosettafold/network/parsers.py
code served (permissive licence) · get_code("760631a5eecfc7cf")
How it ran

no invocation recorded.

write_info_file Not yet run
unverified (harvested, no recorded run)
trachote/predict_elastic_tensor/se3t/utils/utils_logging.py
code served (permissive licence) · get_code("421a0439514724ca")
How it ran

no invocation recorded.

Also present in other papers' code

Of 28 samples lifted from this paper's linked repositories, 28 carry a function identity and 0 do not. Of the 28 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 28 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 introduce the SE(3)-Transformer, a variant of the self-attention module for 3D point clouds and graphs, which is equivariant under continuous 3D roto-translations. Equivariance is important to ensure stable and predictable performance in the presence of nuisance transformations of the data input. A positive corollary of equivariance is increased weight-tying within the model. The SE(3)-Transformer leverages the benefits of self-attention to operate on large point clouds and graphs with varying number of points, while guaranteeing SE(3)-equivariance for robustness. We evaluate our model on a toy N-body particle simulation dataset, showcasing the robustness of the predictions under rotations of the input. We further achieve competitive performance on two real-world datasets, ScanObjectNN and QM9. In all cases, our model outperforms a strong, non-equivariant attention baseline and an equivariant model without attention.

For agents

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

get_harvested_code_for_paper("2006.10503")
get_code_for_paper("2006.10503")
have("2006.10503")

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

curl https://syntology.ai/api/ran/2006.10503.json

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

[![Syntology run record](https://syntology.ai/api/ran/2006.10503.svg)](https://syntology.ai/paper/2006.10503)

Connect an agent — have() is free.