SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2003.03290 · ICLR · 2020

Towards a predictive spatio-temporal representation of brain data

Tiago Azevedo, Luca Passamonti, Pietro Liò, Nicola Toschi

arXiv · PDF · Open in the Atlas

Code that ran

9 of 10 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 9 have no contract verdict; 0 failures of our instrument; 1 no recorded run), measured 2026-11-21; 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/2003.03290.json · badge: /api/ran/2003.03290.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
tjiagoM/spatio-temporal-brain community (classifier label)
classifier basis not recorded on the link
9 ran · 1 no recorded run
FunctionStatusWhere it lives
calculate_indegree_histogram Ran
ran on a synthesized input
tjiagoM/spatio-temporal-brain/utils.py
code served (permissive licence) · get_code("531350504db75dcc")
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 int, shape (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
from torch_geometric.data import Data

torch.manual_seed(0)

# Create a minimal dataset with a few graph objects
# Each graph needs: edge_index and num_nodes attributes
graph1 = Data(
    edge_index=torch.tensor([[0, 1, 2], [1, 2, 0]], dtype=torch.long),
    num_nodes=3
)

graph2 = Data(
    edge_index=torch.tensor([[0, 1], [1, 2]], dtype=torch.long),
    num_nodes=3
)

graph3 = Data(
    edge_index=torch.tensor([[0, 0, 1, 1, 2], [1, 2, 0, 2, 1]], dtype=torch.long),
    num_nodes=3
)

tmp_dataset = [graph1, graph2, graph3]

ARGS = {
    "tmp_dataset": tmp_dataset
}
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 Tensor, ndim 1, dtype family int, shape (3,) — shape (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
create_name_for_flattencorrs_dataset Ran
ran on a synthesized input
tjiagoM/spatio-temporal-brain/utils.py
code served (permissive licence) · get_code("43f68b1db7739827")
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.

from enum import Enum

class DatasetType(Enum):
    type_a = "type_a"
    type_b = "type_b"

class AnalysisType(Enum):
    analysis_x = "analysis_x"
    analysis_y = "analysis_y"

class ParamConnType(Enum):
    conn_full = "conn_full"
    conn_sparse = "conn_sparse"

run_cfg = {
    'dataset_type': DatasetType.type_a,
    'analysis_type': AnalysisType.analysis_x,
    'param_conn_type': ParamConnType.conn_full,
    'num_nodes': 10,
    'time_length': 50
}

ARGS = {
    'run_cfg': run_cfg
}
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
get_desikan_tracks_path Ran
ran on a synthesized input
tjiagoM/spatio-temporal-brain/datasets.py
code served (permissive licence) · get_code("7a1adb0ae6c45871")
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 torch
import numpy as np

# The function takes a person parameter (int) and returns a path string
# It doesn't actually use the person value in any complex way - just formats it into a string
# So we just need a valid integer

ARGS = {
    "person": 42
}
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 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
get_desikan_ts_path Ran
ran on a synthesized input
tjiagoM/spatio-temporal-brain/datasets.py
code served (permissive licence) · get_code("5a0e66a9259a33d2")
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 torch

ARGS = {
    'person': 12345,
    'direction': 'LR'
}
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 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
loss_function_ae Ran
ran on a synthesized input
tjiagoM/spatio-temporal-brain/encoders/encoders.py
code served (permissive licence) · get_code("2aaa4262ac3da55f")
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 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)

# Create small tensors with distinct dimensions for reconstruction and input
# recon_x: reconstructed output from autoencoder
# x: original input
# Using shape (2, 4) where 2 is batch size, 4 is feature dimension

recon_x = torch.randn(2, 4)
x = torch.randn(2, 4)

ARGS = {
    "recon_x": recon_x,
    "x": x,
}
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 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
loss_function_vae Ran
ran on a synthesized input
tjiagoM/spatio-temporal-brain/encoders/encoders.py
code served (permissive licence) · get_code("770c30c5951ba450")
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 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.functional as F

torch.manual_seed(0)

# Create small tensors for VAE loss function
batch_size = 2
latent_dim = 4

# recon_x: reconstructed output from decoder, same shape as x
recon_x = torch.randn(batch_size, latent_dim)

# x: original input
x = torch.randn(batch_size, latent_dim)

# mu: mean from encoder
mu = torch.randn(batch_size, latent_dim)

# logvar: log variance from encoder
logvar = torch.randn(batch_size, latent_dim)

ARGS = {
    "recon_x": recon_x,
    "x": x,
    "mu": mu,
    "logvar": logvar,
}
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 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
merge_y_and_others Ran
ran on a synthesized input
tjiagoM/spatio-temporal-brain/utils.py
code served (permissive licence) · get_code("f2becff4fb07511d")
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 int, 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
from sklearn.preprocessing import LabelEncoder

torch.manual_seed(0)

# Create small tensors for the function
ys = torch.tensor([1, 0], dtype=torch.float32)
indices = torch.tensor([2, 3], dtype=torch.int64)

ARGS = {
    "ys": ys,
    "indices": indices
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
sklearn, torch — 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 int, 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
return_regressor_metrics Ran
ran on a synthesized input
tjiagoM/spatio-temporal-brain/main_loop.py
code served (permissive licence) · get_code("8c1aface120b7221")
How it ran

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

import numpy as np
from sklearn.preprocessing import StandardScaler

# Create small numpy arrays for labels and predictions
np.random.seed(0)
labels = np.random.randn(8).astype(np.float32)
pred_prob = labels + np.random.randn(8) * 0.1  # Correlated predictions

# Create a label scaler
label_scaler = StandardScaler()
labels_scaled = label_scaler.fit_transform(labels.reshape(-1, 1))[:, 0]
pred_prob_scaled = label_scaler.transform(pred_prob.reshape(-1, 1))[:, 0]

# Define loss values
loss_value = 0.5
link_loss_value = 0.3
ent_loss_value = 0.2

ARGS = {
    'labels': labels_scaled,
    'pred_prob': pred_prob_scaled,
    'label_scaler': label_scaler,
    'loss_value': loss_value,
    'link_loss_value': link_loss_value,
    'ent_loss_value': ent_loss_value
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy, sklearn — 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
threshold_adj_array Ran
ran on a synthesized input
tjiagoM/spatio-temporal-brain/datasets.py
code served (permissive licence) · get_code("c2e83b06c6237795")
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 (5, 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 numpy as np

# Create a small adjacency matrix
num_nodes = 5
adj_array = np.random.rand(num_nodes, num_nodes)

# Make it symmetric by taking upper triangle
adj_array = np.triu(adj_array)
adj_array = adj_array + adj_array.T

# Set diagonal to 0 initially (will be set to 1.0 in the function)
np.fill_diagonal(adj_array, 0)

threshold = 30

ARGS = {
    "adj_array": adj_array,
    "threshold": threshold,
    "num_nodes": num_nodes,
}
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 2, dtype family float, shape (5, 5) — shape (5, 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
return_classifier_metrics Not yet run
unverified (harvested, no recorded run)
tjiagoM/spatio-temporal-brain/main_loop.py
code served (permissive licence) · get_code("8f3a7ba54836bda8")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 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 10 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 2026-11-21.

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 characterisation of the brain as a "connectome", in which the connections are represented by correlational values across timeseries and as summary measures derived from graph theory analyses, has been very popular in the last years. However, although this representation has advanced our understanding of the brain function, it may represent an oversimplified model. This is because the typical fMRI datasets are constituted by complex and highly heterogeneous timeseries that vary across space (i.e., location of brain regions). We compare various modelling techniques from deep learning and geometric deep learning to pave the way for future research in effectively leveraging the rich spatial and temporal domains of typical fMRI datasets, as well as of other similar datasets. As a proof-of-concept, we compare our approaches in the homogeneous and publicly available Human Connectome Project (HCP) dataset on a supervised binary classification task. We hope that our methodological advances relative to previous "connectomic" measures can ultimately be clinically and computationally relevant by leading to a more nuanced understanding of the brain dynamics in health and disease. Such understanding of the brain can fundamentally reduce the constant specialised clinical expertise in order to accurately understand brain variability.

For agents

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

get_harvested_code_for_paper("2003.03290")
get_code_for_paper("2003.03290")
have("2003.03290")

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

curl https://syntology.ai/api/ran/2003.03290.json

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

[![Syntology run record](https://syntology.ai/api/ran/2003.03290.svg)](https://syntology.ai/paper/2003.03290)

Connect an agent — have() is free.