SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2211.09224 · 2022

Are we certain it's anomalous?

arXiv · PDF · Open in the Atlas

Code that ran

19 of 24 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 17 have no contract verdict; 2 failures of our instrument; 5 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/2211.09224.json · badge: /api/ran/2211.09224.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
aleflabo/HypAD canonical (classifier label)
classifier basis not recorded on the link
19 ran · 5 no recorded run
FunctionStatusWhere it lives
arsinh Ran
ran on a synthesized input
aleflabo/HypAD/math_.py
code served (permissive licence) · get_code("edde7ea27b3af6d7")
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; 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 takes a single parameter: x, which is a torch.Tensor
# The function applies mathematical operations: addition, sqrt, pow, clamp_min, log, and to
# So x needs to be a tensor with numeric values

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
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 — no output shape is stated: observe_signature.py could not observe it (import:OSError)
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
artanh Ran
ran on a synthesized input
aleflabo/HypAD/math_.py
code served (permissive licence) · get_code("00f30c4d850a7768")
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; 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 artanh function takes a single tensor parameter x
# It performs clamp, log, and arithmetic operations on it
# So we need a small tensor with values in a reasonable range for artanh

x = torch.randn(2, 4) * 0.5  # Values roughly in [-1, 1] range, clamped by function

ARGS = {
    "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 2, dtype family float — no output shape is stated: observe_signature.py could not observe it (import:OSError)
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_accuracy Ran
ran on a synthesized input
aleflabo/HypAD/hyperspace/utils.py
code served (permissive licence) · get_code("d09bfd0a1223d58a")
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

torch.manual_seed(0)

# Create output tensor: (B=2, N=4) - batch size 2, number of classes 4
output = torch.randn(2, 4)

# Create target tensor: (B=2,) - batch size 2
target = torch.tensor([1, 3])

ARGS = {
    "output": output,
    "target": target,
}
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
calc_accuracy_binary Ran
ran; our contract draft was wrong, not the code
aleflabo/HypAD/hyperspace/utils.py
code served (permissive licence) · get_code("8e8878e3cc683e8d")
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

torch.manual_seed(0)

# Create small binary classification tensors
# output: logits before sigmoid, shape (B, N) where B=2 (batch), N=4 (features)
output = torch.randn(2, 4)

# target: binary labels, shape (B, N)
target = torch.randint(0, 2, (2, 4))

ARGS = {
    "output": output,
    "target": target,
}
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
calc_topk_accuracy Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
aleflabo/HypAD/hyperspace/utils.py
code served (permissive licence) · get_code("fa47a4519d4ef5c1")
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 torch

torch.manual_seed(0)

# Create small tensors for testing
batch_size = 2
num_classes = 4

# output: model predictions (batch_size x num_classes)
# needs to support .topk() method, so it's a tensor
output = torch.randn(batch_size, num_classes)

# target: ground truth labels (batch_size,)
# needs to support .size(0) and .view() and .expand_as(), so it's a tensor
target = torch.tensor([1, 3])

# topk: tuple of k values to compute accuracy for
topk = (1, 2)

ARGS = {
    "output": output,
    "target": target,
    "topk": topk,
}
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 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
casas_anomalies Ran
ran on a synthesized input
aleflabo/HypAD/utils/anomaly_detection_utils.py
code served (permissive licence) · get_code("1acd35953c608551")
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 torch
import numpy as np
import pandas as pd

torch.manual_seed(0)

# Analyze the function:
# - y: needs to be reshaped to (y.shape[0] * y.shape[1], -1), so needs at least 2 dimensions
#   Then indexed with y[i] and compared to 1, so numeric values
# - x_index: needs .shape[0] attribute, and indexed with x_index[i], so array-like
#   Should have same length as flattened y

# Create y: 2D array that will be reshaped
# shape (2, 3) will flatten to (6,) - make it have some 1s and 0s
y = np.array([[0, 1, 1],
              [0, 1, 0]], dtype=np.int32)

# Create x_index: 1D array with length matching flattened y
# y will be reshaped to (6, 1) then sliced to x_index.shape[0]
x_index = np.array([0, 10, 20, 30, 40, 50], dtype=np.float32)

ARGS = {
    'y': y,
    'x_index': x_index
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas, torch — 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
compute_critic_scores Ran
ran on a synthesized input
aleflabo/HypAD/utils/anomaly_detection_utils.py
code served (permissive licence) · get_code("289f57bf7e900dab")
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 (7,) 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 math
import os
import pickle
import numpy as np
import pandas as pd
from scipy import integrate, stats
import tempfile

# Create a temporary directory for pickle files
temp_dir = tempfile.mkdtemp()

# Create minimal test data
rec_scores = np.random.rand(10)  # Shape (10,)
critic_score = np.random.rand(5)  # Shape (5,)
true_signal = np.random.rand(5, 3)  # Shape (5, 3) - 5 time steps, 3 features

# Create a minimal params object
class Params:
    def __init__(self, load=False):
        self.load = load

params = Params(load=False)
path = temp_dir + '/'

# Ensure directory exists
os.makedirs(path, exist_ok=True)

ARGS = {
    'rec_scores': rec_scores,
    'critic_score': critic_score,
    'true_signal': true_signal,
    'params': params,
    'path': path
}
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 ndarray, ndim 1, dtype family float, shape (7,) — shape (7,) 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
compute_supervised_loss Ran
ran on a synthesized input
aleflabo/HypAD/hyperspace/losses.py
code served (permissive licence) · get_code("3b4149036dae58ef")
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 torch
import numpy as np

torch.manual_seed(0)

class Args:
    def __init__(self):
        self.pred_future = False
        self.num_seq = 3
        self.device = 'cpu'
        self.hierarchical_labels = False
        self.early_action = False
        self.dataset = 'test_dataset'

# Global hierarchy shape dict that the function references
sh = {
    'test_dataset': (10, [4, 3, 2])  # (num_classes, hierarchy_levels)
}

# Inject sh into the function's globals for execution
import sys
current_module = sys.modules[__name__]
current_module.sh = sh

# Setup arguments
args = Args()

# Case 1: Simple prediction without hierarchical labels
# pred shape: (B, num_classes) where B=2, num_classes=5
# labels shape: (B,) where B=2
B = 2
pred = torch.randn(B, 5)
labels = torch.tensor([0, 2], dtype=torch.long)

ARGS = {
    'args': args,
    'pred': pred,
    'labels': labels,
    'B': B
}
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 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
convert_date Ran
ran on a synthesized input
aleflabo/HypAD/utils/utils.py
code served (permissive licence) · get_code("e6ee59bdde3e7f5d")
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.

from datetime import datetime
import time

# Create a list of valid Unix timestamps
# Using small, recent timestamps (seconds since epoch)
timelist = [
    1609459200,  # 2021-01-01 00:00:00 UTC
    1640995200,  # 2022-01-01 00:00:00 UTC
]

ARGS = {
    "timelist": timelist
}
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 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
convert_date_single Ran
ran on a synthesized input
aleflabo/HypAD/utils/anomaly_detection_utils.py
code served (permissive licence) · get_code("6c0b8fb7e97280e9")
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 datetime. 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 datetime import datetime

# The function convert_date_single takes a single parameter x
# and calls datetime.fromtimestamp(x)
# x should be a numeric timestamp (int or float)

# A valid Unix timestamp (e.g., January 1, 2020)
timestamp_value = 1577836800

ARGS = {
    "x": timestamp_value
}
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 datetime — a datetime output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a datetime 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
download Ran
ran on a synthesized input
aleflabo/HypAD/utils/data.py
code served (permissive licence) · get_code("8cd82307c19e9ebc")
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 logging
import os
import pandas as pd
import tempfile
import shutil

# Set up a temporary directory for data_path to avoid actual file I/O
temp_dir = tempfile.mkdtemp()

# Create mock CSV files in the temp directory
mock_data = pd.DataFrame({
    'col1': [1, 2, 3, 4, 5],
    'col2': ['a', 'b', 'c', 'd', 'e'],
    'col3': [1.1, 2.2, 3.3, 4.4, 5.5]
})
mock_data.to_csv(os.path.join(temp_dir, 'sample_data.csv'), index=False)

ARGS = {
    'name': 'sample_data',
    'test_size': None,
    'data_path': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
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
load_csv Ran
ran on a synthesized input
aleflabo/HypAD/utils/data.py
code served (permissive licence) · get_code("7996920fef9af9d0")
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 tempfile
import os

# Create a temporary CSV file for testing
temp_dir = tempfile.mkdtemp()
csv_path = os.path.join(temp_dir, "test_data.csv")

# Write a simple CSV file with headers
csv_content = """timestamp,value,other_col
2024-01-01,10.5,A
2024-01-02,20.3,B
2024-01-03,15.7,C
"""

with open(csv_path, 'w') as f:
    f.write(csv_content)

# Define ARGS with all parameters
ARGS = {
    'path': csv_path,
    'timestamp_column': 0,
    'value_column': 1,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
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
pairwise_distances Ran
ran on a synthesized input
aleflabo/HypAD/hyperspace/poincare_distance.py
code served (permissive licence) · get_code("cd051a881427eb3f")
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, 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
import numpy as np

torch.manual_seed(0)

# Create small test tensors
# x: NxD matrix (N=2 samples, D=4 dimensions)
x = torch.randn(2, 4)

# y: MxD matrix (M=3 samples, D=4 dimensions) - distinct size to test the y parameter
y = torch.randn(3, 4)

ARGS = {
    "x": x,
    "y": y,
}
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 Tensor, ndim 2, dtype family float, 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
poincare_distance Ran
ran on a synthesized input
aleflabo/HypAD/hyperspace/poincare_distance.py
code served (permissive licence) · get_code("5c19852209e2c5fe")
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, 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
import numpy as np

torch.manual_seed(0)

# Create small tensors for pred and gt
# pred: N_pred x D, gt: N_gt x D
# Using distinct sizes: N_pred=2, N_gt=3, D=4
pred = torch.randn(2, 4)
gt = torch.randn(3, 4)

ARGS = {
    "pred": pred,
    "gt": gt,
}
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 Tensor, ndim 2, dtype family float, 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
square_norm Ran
ran on a synthesized input
aleflabo/HypAD/hyperspace/poincare_distance.py
code served (permissive licence) · get_code("3f7e8409c9f2f2ba")
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 float, 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

torch.manual_seed(0)

# The function takes a single parameter 'x' which is a tensor
# It computes the norm along the last dimension (dim=-1) with p=2
# So x should be a tensor with at least 1 dimension

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
torch — 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 float, 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
tanh Ran
ran on a synthesized input
aleflabo/HypAD/math_.py
code served (permissive licence) · get_code("c33c4cbe2f422504")
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; 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 takes a single parameter 'x' which should be a tensor
# It calls .clamp() and .tanh() on it, so it needs to be a torch.Tensor
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
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 — no output shape is stated: observe_signature.py could not observe it (import:OSError)
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
unroll_signal Ran
ran on a synthesized input
aleflabo/HypAD/models/tadgan.py
code served (permissive licence) · get_code("f8165a0a30ee6414")
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 float32, 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 numpy as np

class MockModel:
    def unroll_signal(self, x):
        x = np.array(x).reshape(100)
        return np.median(x)

# Create a small stand-in object
model = MockModel()

# Create input data: reshape to 100 elements
x_input = np.arange(100, dtype=np.float32)

ARGS = {
    'self': model,
    'x': x_input
}
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 float32, 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
unroll_ts Ran
ran on a synthesized input
aleflabo/HypAD/utils/utils.py
code served (permissive licence) · get_code("7c00efad4a3148c1")
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

# Create a small 2D numpy array to represent y_hat
# Shape: (batch_size, pred_length) = (4, 3)
# The function iterates through shape[0] and shape[1], so we need both dimensions
y_hat = 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]
], dtype=np.float32)

ARGS = {
    "y_hat": y_hat
}
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
yahoo_preprocess Ran
ran on a synthesized input
aleflabo/HypAD/utils/dataloader.py
code served (permissive licence) · get_code("3e220c20c168a267")
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
from datetime import datetime

# Create a sample DataFrame that matches what yahoo_preprocess expects
df = pd.DataFrame({
    'value': np.array([1.0, 2.0, 3.0, 4.0, 5.0]),
    'is_anomaly': np.array([0, 0, 1, 0, 0]),
    'timestamp': np.array([1.0, 2.0, 3.0, 4.0, 5.0])
})

ARGS = {
    'df': df
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
dateutil, numpy, pandas, scipy — 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
load_signal Not yet run
unverified (harvested, no recorded run)
aleflabo/HypAD/utils/data.py
code served (permissive licence) · get_code("068dd3957754e24a")
How it ran

no invocation recorded.

mobius_linear Not yet run
unverified (harvested, no recorded run)
aleflabo/HypAD/hyperspace/hyrnn_nets.py
code served (permissive licence) · get_code("a4ee074610449d21")
How it ran

no invocation recorded.

mobius_matvec Not yet run
unverified (harvested, no recorded run)
aleflabo/HypAD/hyperspace/hyrnn_nets.py
code served (permissive licence) · get_code("9dffae8f52f9f0a0")
How it ran

no invocation recorded.

one_rnn_transform Not yet run
unverified (harvested, no recorded run)
aleflabo/HypAD/hyperspace/hyrnn_nets.py
code served (permissive licence) · get_code("9f339b3c440a606a")
How it ran

no invocation recorded.

save_known_anomalies Not yet run
unverified (harvested, no recorded run)
aleflabo/HypAD/utils/dataloader.py
code served (permissive licence) · get_code("00241d35d68123e3")
How it ran

no invocation recorded.

Also present in other papers' code

Of 24 samples lifted from this paper's linked repositories, 24 carry a function identity and 0 do not. Of the 24 distinct functions with an identity, 5 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 1 later) and 19 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.

FunctionAlso present inEarliest-datedOthers named
calc_accuracy 2 other papers
2 dated earlier
1909.04656 (2019-09)
Video Representation Learning by Dense Predictive Coding
2008.01065 (2020-08)
Memory-augmented Dense Predictive Coding for Video Representation Learning
mobius_matvec 2 other papers
1 dated earlier, 1 later
2101.01600 (2021-01)
Learning the Predictability of the Future
2303.06242 (2023-03)
HYperbolic Self-Paced Learning for Self-Supervised Skeleton-based Action Representations
calc_accuracy_binary 1 other paper
1 dated earlier
1909.04656 (2019-09)
Video Representation Learning by Dense Predictive Coding
calc_topk_accuracy 1 other paper
1 dated earlier
1909.04656 (2019-09)
Video Representation Learning by Dense Predictive Coding
mobius_linear 1 other paper
1 dated earlier
2101.01600 (2021-01)
Learning the Predictability of the Future

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

The progress in modelling time series and, more generally, sequences of structured data has recently revamped research in anomaly detection. The task stands for identifying abnormal behaviors in financial series, IT systems, aerospace measurements, and the medical domain, where anomaly detection may aid in isolating cases of depression and attend the elderly. Anomaly detection in time series is a complex task since anomalies are rare due to highly non-linear temporal correlations and since the definition of anomalous is sometimes subjective. Here we propose the novel use of Hyperbolic uncertainty for Anomaly Detection (HypAD). HypAD learns self-supervisedly to reconstruct the input signal. We adopt best practices from the state-of-the-art to encode the sequence by an LSTM, jointly learned with a decoder to reconstruct the signal, with the aid of GAN critics. Uncertainty is estimated end-to-end by means of a hyperbolic neural network. By using uncertainty, HypAD may assess whether it is certain about the input signal but it fails to reconstruct it because this is anomalous; or whether the reconstruction error does not necessarily imply anomaly, as the model is uncertain, e.g. a complex but regular input signal. The novel key idea is that a detectable anomaly is one where the model is certain but it predicts wrongly. HypAD outperforms the current state-of-the-art for univariate anomaly detection on established benchmarks based on data from NASA, Yahoo, Numenta, Amazon, and Twitter. It also yields state-of-the-art performance on a multivariate dataset of anomaly activities in elderly home residences, and it outperforms the baseline on SWaT. Overall, HypAD yields the lowest false alarms at the best performance rate, thanks to successfully identifying detectable anomalies.

For agents

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

get_harvested_code_for_paper("2211.09224")
get_code_for_paper("2211.09224")
have("2211.09224")

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

curl https://syntology.ai/api/ran/2211.09224.json

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

[![Syntology run record](https://syntology.ai/api/ran/2211.09224.svg)](https://syntology.ai/paper/2211.09224)

Connect an agent — have() is free.