SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.01433 · 2026

DCD: Decomposition-based Causal Discovery from Autocorrelated and Non-Stationary Temporal Data

Muhammad Ferdous, Osman Gani

arXiv · PDF · Open in the Atlas

Code that ran

14 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 13 failures of our instrument; 0 no recorded run), measured 2026-11-21; 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/2602.01433.json · badge: /api/ran/2602.01433.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
noname2122/DCD-TMLR-2025 canonical (classifier label)
classifier basis: census
14 ran · 0 no recorded run
FunctionStatusWhere it lives
calculate_metrics Ran
ran; our contract draft was wrong, not the code
noname2122/DCD-TMLR-2025/experiments/run_dynotears_baseline.py
code served (permissive licence) · get_code("9500b6a629b1e792")
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)

# Define edges as sets of tuples representing graph edges
# Each edge is a tuple of two node identifiers
discovered_edges = [(0, 1), (1, 2), (2, 3), (0, 3)]
ground_truth_edges = [(0, 1), (1, 2), (3, 0), (1, 3)]

ARGS = {
    "discovered_edges": discovered_edges,
    "ground_truth_edges": ground_truth_edges,
}
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
convert_ipynb_to_py Ran
ran; our contract draft was wrong, not the code
noname2122/DCD-TMLR-2025/src/dcd/utils/extract_code.py
code served (permissive licence) · get_code("d3ac1ff2e7cf7051")
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 PosixPath. 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 json
from pathlib import Path
import tempfile

# Create a temporary directory and a sample notebook file
temp_dir = Path(tempfile.mkdtemp())

# Create a minimal valid Jupyter notebook structure
notebook_content = {
    "cells": [
        {
            "cell_type": "code",
            "source": ["import numpy as np\n", "x = np.array([1, 2, 3])"]
        },
        {
            "cell_type": "markdown",
            "source": ["# This is a markdown cell"]
        },
        {
            "cell_type": "code",
            "source": ["print(x)"]
        }
    ],
    "metadata": {},
    "nbformat": 4,
    "nbformat_minor": 2
}

# Write the notebook to a temporary file
ipynb_path = temp_dir / "test_notebook.ipynb"
with ipynb_path.open('w', encoding='utf-8') as f:
    json.dump(notebook_content, f)

ARGS = {
    "ipynb_path": ipynb_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 PosixPath — a PosixPath output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a PosixPath 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
decompose_all_fixed Ran
ran; our contract draft was wrong, not the code
noname2122/DCD-TMLR-2025/experiments/ablation.py
code served (permissive licence) · get_code("e2632cdaa60f375b")
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 DataFrame. 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 pandas as pd
import numpy as np

# Create a small time series dataframe
np.random.seed(0)
n_samples = 100
time_index = np.arange(n_samples)

# Create synthetic time series data with multiple features
df = pd.DataFrame({
    'time': time_index,
    'feature_1': np.sin(np.arange(n_samples) * 0.1) + np.random.normal(0, 0.1, n_samples),
    'feature_2': np.cos(np.arange(n_samples) * 0.05) + np.random.normal(0, 0.1, n_samples),
    'feature_3': np.random.normal(0, 1, n_samples)
})

period = 12

ARGS = {
    'df': df,
    'period': period
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas, statsmodels — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
decompose_all_fixed Ran
ran; our contract draft was wrong, not the code
noname2122/DCD-TMLR-2025/experiments/extensive_ablation.py
code served (permissive licence) · get_code("88225f58b70d9180")
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 DataFrame. 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 statsmodels.tsa.seasonal import STL

# Create a small sample dataframe with a 'time' column and feature columns
np.random.seed(0)
time_values = np.arange(0, 100)
feature_1 = np.sin(np.arange(0, 100) * 0.1) + np.random.normal(0, 0.1, 100)
feature_2 = np.cos(np.arange(0, 100) * 0.05) + np.random.normal(0, 0.1, 100)

df = pd.DataFrame({
    'time': time_values,
    'feature_1': feature_1,
    'feature_2': feature_2
})

# Set a reasonable period (must be > 2 for STL to work properly)
period = 12

ARGS = {
    'df': df,
    'period': period
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas, statsmodels — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
decompose_residual Ran
ran; our contract draft was wrong, not the code
noname2122/DCD-TMLR-2025/experiments/replicate_paper_tables.py
code served (permissive licence) · get_code("2bf4babc61bf90c3")
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 Series. 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 statsmodels.tsa.seasonal import STL

# Create a simple time series as a pandas Series
np.random.seed(0)
time_series_values = np.sin(np.linspace(0, 4 * np.pi, 100)) + np.random.normal(0, 0.1, 100)
series = pd.Series(time_series_values)

# Define ARGS dictionary with valid arguments
ARGS = {
    "series": series,
    "period": 12
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas, statsmodels — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Series — a Series output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Series output the recorded ndim 1 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
decompose_series Ran
ran; our contract draft was wrong, not the code
noname2122/DCD-TMLR-2025/src/dcd/core.py
code served (permissive licence) · get_code("da8c86b330c7427d")
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
from statsmodels.tsa.seasonal import STL

# Create a simple time series with seasonal pattern
np.random.seed(0)
n_points = 100
series = np.sin(np.linspace(0, 8*np.pi, n_points)) + np.random.normal(0, 0.1, n_points)

# Define arguments
ARGS = {
    "series": series,
    "period": 12
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, statsmodels — 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
decompose_series Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
noname2122/DCD-TMLR-2025/experiments/ablation.py
code served (permissive licence) · get_code("12337eab296e912f")
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 (100,) 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
import torch

torch.manual_seed(0)

# The function expects:
# - series: a time series object that STL can work with (numpy array or pandas Series)
# - period: an integer or None

# Create a simple time series using numpy
series = np.random.randn(100) + np.sin(np.arange(100) * 2 * np.pi / 12)

# Define period as an integer
period = 12

ARGS = {
    "series": series,
    "period": period,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, statsmodels, 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 float, shape (100,) — shape (100,) 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
decompose_series Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
noname2122/DCD-TMLR-2025/experiments/extensive_ablation.py
code served (permissive licence) · get_code("1605dd8270713b8b")
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 (30,) 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

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

# Create a simple time series
series = np.array([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 
                   11.0, 12.0, 13.0, 14.0, 15.0, 16.0, 17.0, 18.0, 19.0, 20.0,
                   21.0, 22.0, 23.0, 24.0, 25.0, 26.0, 27.0, 28.0, 29.0, 30.0])

period = 12

ARGS = {
    "series": series,
    "period": period
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, statsmodels — 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 (30,) — shape (30,) 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
find_optimal_period Ran
ran; our contract draft was wrong, not the code
noname2122/DCD-TMLR-2025/experiments/replicate_paper_tables.py
code served (permissive licence) · get_code("d90046ec34fc22bc")
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 int, 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 numpy as np
import pandas as pd

# Create a simple time series with a seasonal pattern
np.random.seed(0)
n_samples = 200
t = np.arange(n_samples)
# Create a series with a dominant period of around 12
seasonal_component = 10 * np.sin(2 * np.pi * t / 12)
trend_component = 0.5 * t
noise = np.random.normal(0, 1, n_samples)
series_values = seasonal_component + trend_component + noise

series = pd.Series(series_values)

ARGS = {
    "series": series,
    "max_period": 40,
    "power_threshold": 0.05
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, 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
find_optimal_stl_period Ran
ran; our contract draft was wrong, not the code
noname2122/DCD-TMLR-2025/src/dcd/core.py
code served (permissive licence) · get_code("e269b84daada4f2a")
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 int, 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 numpy as np
from statsmodels.tsa.seasonal import STL

# Create a synthetic time series for testing
np.random.seed(42)
series = np.sin(np.linspace(0, 4 * np.pi, 100)) + np.random.normal(0, 0.1, 100)

ARGS = {
    "series": series,
    "max_period": 50,
    "variance_threshold": 0.1,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, statsmodels — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, 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
generate_synthetic_data Ran
ran, honoured the contract we drafted
noname2122/DCD-TMLR-2025/experiments/ablation.py
code served (permissive licence) · get_code("7735da1885fe556d")
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 DataFrame. 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

ARGS = {
    'n_samples': 500
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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_data_and_edges Ran
ran; our contract draft was wrong, not the code
noname2122/DCD-TMLR-2025/experiments/extensive_ablation.py
code served (permissive licence) · get_code("be642c1e6d0fab50")
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

# Set random seed for reproducibility
np.random.seed(45)

# Define ARGS dictionary with valid arguments for get_data_and_edges
ARGS = {
    'n_vars': 4,
    'n_samples': 100
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — 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
load_dataset Ran
ran; our contract draft was wrong, not the code
noname2122/DCD-TMLR-2025/src/dcd/core.py
code served (permissive licence) · get_code("8a12e24b4f992edf")
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 DataFrame. 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 pandas as pd
import numpy as np

# Create a sample dataframe with numeric columns
df = pd.DataFrame({
    'time': [0, 1, 2, 3],
    'feature1': [1.0, 2.0, 3.0, 4.0],
    'feature2': [5.0, 6.0, 7.0, 8.0],
    'feature3': [9.0, 10.0, 11.0, 12.0]
})

ARGS = {
    'df': df,
    'time_col': 'time',
    'feature_cols': ['feature1', 'feature2', 'feature3']
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
stl_decompose_df Ran
ran; our contract draft was wrong, not the code
noname2122/DCD-TMLR-2025/experiments/replicate_paper_tables.py
code served (permissive licence) · get_code("a1dc19effe62c003")
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 DataFrame. 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

# Create a small DataFrame with time series data
np.random.seed(0)
n_samples = 100
time_index = np.arange(n_samples)

# Create synthetic time series with some seasonal patterns
t = np.arange(n_samples, dtype=float)
feature1 = 10 + 2 * np.sin(2 * np.pi * t / 12) + 0.1 * t + np.random.normal(0, 0.5, n_samples)
feature2 = 20 + 3 * np.sin(2 * np.pi * t / 8) + 0.05 * t + np.random.normal(0, 0.5, n_samples)
feature3 = 15 + np.sin(2 * np.pi * t / 6) + np.random.normal(0, 0.3, n_samples)

df = pd.DataFrame({
    'time': time_index,
    'feature1': feature1,
    'feature2': feature2,
    'feature3': feature3
})

ARGS = {
    'df': df,
    'time_col': 'time'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas, scipy, statsmodels — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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

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 13 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 13 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

Multivariate time series in domains such as finance, climate science, and healthcare often exhibit long-term trends, seasonal patterns, and short-term fluctuations, which complicate causal inference under non-stationarity and autocorrelation. Most causal discovery methods operate on raw observations and are therefore susceptible to spurious edges and misattributed temporal dependencies. We introduce a decomposition-based causal discovery (DCD) framework that separates each series into trend, seasonal, and residual components and performs componentspecific causal analysis: stationarity tests for trends, kernel-based dependence tests for seasonality, and constraint-based causal discovery for residuals. The resulting component-level graphs are integrated into a unified multi-scale causal structure that distinguishes time-proxy edges (exogenous temporal drivers) from mechanistic edges (variable-to-variable dependence). We establish identifiability guarantees under spectral separability and bounded leakage, and quantify the structural error in terms of a leakage parameter ε and the Type-I error of the downstream causal test. Across synthetic benchmarks and real-world climate and electricity datasets, DCD recovers ground-truth causal structure more accurately than state-of-the-art baselines under strong non-stationarity and autocorrelation, and degrades gracefully when spectral separability is partially violated. Code is available at https://github.com/noname2122/DCD-TMLR-2025.

For agents

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

get_harvested_code_for_paper("2602.01433")
get_code_for_paper("2602.01433")
have("2602.01433")

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

curl https://syntology.ai/api/ran/2602.01433.json

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

[![Syntology run record](https://syntology.ai/api/ran/2602.01433.svg)](https://syntology.ai/paper/2602.01433)

Connect an agent — have() is free.