SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.07606 · 2020

timeXplain -- A Framework for Explaining the Predictions of Time Series Classifiers

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 2026-11-22; 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/2007.07606.json · badge: /api/ran/2007.07606.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
loadingbyte/timexplain canonical (classifier label)
classifier basis: census
9 ran · 1 no recorded run
FunctionStatusWhere it lives
classname Ran
ran on a synthesized input
loadingbyte/timexplain/timexplain/_utils.py
code served (permissive licence) · get_code("eeae61a6df0f59b7")
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

def classname(obj):
    cls = type(obj)
    return f'{cls.__module__}.{cls.__qualname__}'

# The function only calls type() on obj and accesses __module__ and __qualname__
# We can pass any Python object. Let's use a simple tensor as a test object.

torch.manual_seed(0)
test_tensor = torch.randn(2, 4)

ARGS = {
    'obj': test_tensor
}
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
ellip_filter Ran
ran on a synthesized input
loadingbyte/timexplain/timexplain/om/_freqdice.py
code served (permissive licence) · get_code("941e9be5fd59c2a1")
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. 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
from scipy.signal import iirdesign, firls, filtfilt, sosfiltfilt
from typing import Sequence, Tuple

torch.manual_seed(0)

# The function ellip_filter takes:
# - size_x: int - the size/length parameter used for Nyquist frequency calculation
# - stopbands: Sequence[Tuple[float, float]] - a sequence of (f_lower, f_upper) tuples

# Based on the function body:
# - size_x is used as an int for comparisons and division (size_x // 2)
# - stopbands is iterated over to get (f_lower, f_upper) tuples
# - The frequencies need to be valid normalized frequencies relative to size_x

# Create valid test arguments
size_x_val = 100  # Sample size - provides reasonable Nyquist frequency (50)

# Create stopbands with frequencies that are reasonable
# Frequencies should be between 0 and size_x//2 (Nyquist frequency)
# Let's use frequencies that won't trigger both lowcut and highcut at once
stopbands_val = [
    (10.0, 20.0),   # Mid-range stopband
    (30.0, 40.0),   # Another mid-range stopband
]

ARGS = {
    'size_x': size_x_val,
    'stopbands': stopbands_val,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list — a list 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
firls_filter Ran
ran on a synthesized input
loadingbyte/timexplain/timexplain/om/_freqdice.py
code served (permissive licence) · get_code("412b65f1aa7abd97")
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. 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 scipy.signal import firls

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

# Define ARGS dict with all parameters for firls_filter
ARGS = {
    'size_x': 128,
    'stopbands': [(10.0, 20.0), (40.0, 50.0)]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list — a list 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
stats_global Ran
ran on a synthesized input
loadingbyte/timexplain/timexplain/om/_statistic.py
code served (permissive licence) · get_code("b80a115b1f59553e")
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 Stats. 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

# Construct X_bg as a small numpy array
# The function calls np.asarray(X_bg) and np.std(np.asarray(X_bg), axis=1)
# So X_bg should be 2D array-like to compute std along axis 1
X_bg = np.array([
    [1.0, 2.0, 3.0, 4.0],
    [5.0, 6.0, 7.0, 8.0]
], dtype=np.float32)

ARGS = {
    "X_bg": X_bg
}
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 Stats — a Stats output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Stats 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
stats_sample Ran
ran on a synthesized input
loadingbyte/timexplain/timexplain/om/_statistic.py
code served (permissive licence) · get_code("710c6b6b5c7cc0e8")
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 Stats. 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 typing import Union, Optional, Callable, NamedTuple, Sequence

class Stats(NamedTuple):
    mean: float
    stddev: float

def _sample(X_bg):
    assert X_bg is not None
    X_bg = np.asarray(X_bg)
    return X_bg[np.random.randint(X_bg.shape[0])]

def stats_sample(X_bg) -> Stats:
    return Stats(np.mean(_sample(X_bg)), np.std(_sample(X_bg)))

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

# Create sample background data
# Shape: (5, 4) - 5 samples, 4 features
X_bg = np.array([
    [1.0, 2.0, 3.0, 4.0],
    [2.0, 3.0, 4.0, 5.0],
    [3.0, 4.0, 5.0, 6.0],
    [4.0, 5.0, 6.0, 7.0],
    [5.0, 6.0, 7.0, 8.0]
])

ARGS = {
    "X_bg": X_bg
}
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 Stats — a Stats output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Stats 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
take_if_deep Ran
ran on a synthesized input
loadingbyte/timexplain/timexplain/_utils.py
code served (permissive licence) · get_code("9b9c5eb3e05f686f")
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 int, shape (2, 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
from scipy import sparse

# Test case 1: ndarray with ndim >= ndim parameter and idx provided
arr1 = np.array([[1, 2, 3], [4, 5, 6]])  # 2D array
idx1 = 0
ndim1 = 3
error1 = "Array is not deep enough"

# Test case 2: sparse matrix with ndim >= ndim parameter and single row (idx=None should work)
arr2 = sparse.csr_matrix(np.array([[1, 2, 3, 4]]))  # 1x4 sparse matrix
idx2 = None
ndim2 = 2
error2 = "Cannot index sparse matrix without idx"

# Test case 3: list converted to ndarray with ndim < ndim parameter (should return as-is)
arr3 = [1, 2, 3, 4]  # 1D list
idx3 = None
ndim3 = 2
error3 = "Not deep enough"

ARGS = {
    "arr": arr1,
    "idx": idx1,
    "ndim": ndim1,
    "error": error1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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 int, shape (2, 3) — shape (2, 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
x_global_mean Ran
ran on a synthesized input
loadingbyte/timexplain/timexplain/om/_surrogate.py
code served (permissive licence) · get_code("613705235f8996c8")
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 (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 numpy as np

# x_global_mean takes X_bg as input
# It converts X_bg to a numpy array, computes the global mean, and repeats it X_bg.shape[1] times
# X_bg needs to be array-like with at least 2 dimensions (accessed via .shape[1])

X_bg = np.array([
    [1.0, 2.0, 3.0, 4.0],
    [5.0, 6.0, 7.0, 8.0]
])

ARGS = {
    "X_bg": X_bg
}
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 (4,) — shape (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
x_local_mean Ran
ran on a synthesized input
loadingbyte/timexplain/timexplain/om/_surrogate.py
code served (permissive licence) · get_code("44814403dd425a00")
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 (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 numpy as np

# X_bg is used with np.asarray() and then np.mean() with axis=0
# It should be array-like with shape (batch, features)
# Using distinct sizes: batch=2, features=4

X_bg = np.array([
    [1.0, 2.0, 3.0, 4.0],
    [5.0, 6.0, 7.0, 8.0]
])

ARGS = {
    "X_bg": X_bg
}
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 (4,) — shape (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
x_zero Ran
ran on a synthesized input
loadingbyte/timexplain/timexplain/om/_surrogate.py
code served (permissive licence) · get_code("22dc74709b0cfc96")
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 (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 numpy as np

# X_bg is used with:
# - np.shape(X_bg) to get shape
# - indexing [1] on the shape to get second dimension
# So X_bg should be a 2D array-like object

X_bg = np.random.randn(2, 4)

ARGS = {
    "X_bg": X_bg
}
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 (4,) — shape (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
is_iterable Not yet run
unverified (harvested, no recorded run)
loadingbyte/timexplain/timexplain/_utils.py
code served (permissive licence) · get_code("8a166ad6dd2c90b6")
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 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Modern time series classifiers display impressive predictive capabilities, yet their decision-making processes mostly remain black boxes to the user. At the same time, model-agnostic explainers, such as the recently proposed SHAP, promise to make the predictions of machine learning models interpretable, provided there are well-designed domain mappings. We bring both worlds together in our timeXplain framework, extending the reach of explainable artificial intelligence to time series classification and value prediction. We present novel domain mappings for the time domain, frequency domain, and time series statistics and analyze their explicative power as well as their limits. We employ a novel evaluation metric to experimentally compare timeXplain to several model-specific explanation approaches for state-of-the-art time series classifiers.

For agents

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

get_harvested_code_for_paper("2007.07606")
get_code_for_paper("2007.07606")
have("2007.07606")

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

curl https://syntology.ai/api/ran/2007.07606.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.07606.svg)](https://syntology.ai/paper/2007.07606)

Connect an agent — have() is free.