SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2509.04449 · NeurIPS · 2025

CHRONOGRAPH: A Real-World Graph-Based Multivariate Time Series Dataset

Elena Burceanu, Andrei Manolache, Adrian Lutu, Ioana Pintilie

arXiv · PDF · Open in the Atlas

Code that ran

6 of 9 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 4 failures of our instrument; 3 no recorded run), measured 2026-11-22; 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/2509.04449.json · badge: /api/ran/2509.04449.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
bit-ml/ChronoGraph canonical (classifier label)
classifier basis: census
6 ran · 3 no recorded run
FunctionStatusWhere it lives
eval_vae Ran
ran; our contract draft was wrong, not the code
bit-ml/ChronoGraph/code/autoencoder.py
code served (permissive licence) · get_code("fa0e61195fb0f260")
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 torch.nn as nn

torch.manual_seed(0)

class SimpleVAE(nn.Module):
    """Minimal VAE model that returns (x_recon, mu, logvar)"""
    def __init__(self, input_dim=4, latent_dim=2):
        super().__init__()
        self.encoder = nn.Linear(input_dim, latent_dim * 2)
        self.decoder = nn.Linear(latent_dim, input_dim)
        self.latent_dim = latent_dim
    
    def forward(self, x):
        # Encode to mu and logvar
        encoded = self.encoder(x)
        mu = encoded[:, :self.latent_dim]
        logvar = encoded[:, self.latent_dim:]
        
        # Reparameterization trick
        std = torch.exp(0.5 * logvar)
        eps = torch.randn_like(std)
        z = mu + eps * std
        
        # Decode
        x_recon = self.decoder(z)
        
        return x_recon, mu, logvar

# Create a simple VAE model
model = SimpleVAE(input_dim=4, latent_dim=2)
model.eval()

# Create test data: batch_size=2, feature_dim=4
x = torch.randn(2, 4)

ARGS = {
    "model": model,
    "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 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
false_neg Ran
ran, honoured the contract we drafted
bit-ml/ChronoGraph/code/metrics.py
code served (permissive licence) · get_code("20f37c53c7b2ed98")
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 float64, ndim 0, 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

# Construct valid arguments for false_neg function
# The function expects:
# - preds: array-like of predictions (0 or 1)
# - labels: array-like of ground truth labels (0 or 1)

# Create small numpy arrays with binary values
preds = np.array([0, 1, 1, 0, 1, 1])
labels = np.array([0, 1, 0, 0, 1, 0])

ARGS = {
    "preds": preds,
    "labels": labels
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
false_pos Ran
ran, honoured the contract we drafted
bit-ml/ChronoGraph/code/metrics.py
code served (permissive licence) · get_code("8c9cfb742a5a3649")
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 float64, ndim 0, 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

# Construct small numpy arrays for predictions and labels
# preds: binary predictions (0 or 1)
# labels: binary ground truth labels (0 or 1)
preds = np.array([1, 0, 1, 1, 0, 0])
labels = np.array([0, 0, 1, 0, 1, 0])

ARGS = {
    "preds": preds,
    "labels": labels
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
iso_forest Ran
ran; our contract draft was wrong, not the code
bit-ml/ChronoGraph/code/iso_forest.py
code served (permissive licence) · get_code("334bf1bc1acf5c16")
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
from sklearn.ensemble import IsolationForest

np.random.seed(0)

# Create a small dataset with shape [features, samples]
# The function will transpose it to [samples, features]
# Using 4 features and 10 samples
data = np.random.randn(4, 10).astype(np.float32)

# Train/test split percentage
train_percent = 0.7

# Number of estimators
n_estimators = 100

ARGS = {
    'data': data,
    'train_percent': train_percent,
    'n_estimators': n_estimators,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
ocsvm Ran
ran; our contract draft was wrong, not the code
bit-ml/ChronoGraph/code/ocsvm.py
code served (permissive licence) · get_code("683d3e0dfb85dfdc")
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
from sklearn.svm import OneClassSVM

# Create a small synthetic dataset
# The function expects data where it will be transposed
# After transpose, it should have shape (n_samples, n_features)
# Let's create data with shape (n_features, n_samples) before transpose
np.random.seed(0)
n_features = 4
n_samples = 10
data = np.random.randn(n_features, n_samples).astype(np.float32)

# train_percent should be a float between 0 and 1
train_percent = 0.6

ARGS = {
    'data': data,
    'train_percent': train_percent
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
scale Ran
ran; our contract draft was wrong, not the code
bit-ml/ChronoGraph/code/metrics.py
code served (permissive licence) · get_code("a5cf287f94ffc817")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 3, dtype family float, shape (2, 3, 5) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create a small feature array
# Shape: (2, 3, 5) - batch=2, features=3, sequence/time=5
feats = np.random.randn(2, 3, 5).astype(np.float32)

# Define the arguments dictionary
ARGS = {
    'feats': feats,
    'train_proc': 0.6
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family float, shape (2, 3, 5) — shape (2, 3, 5) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_fp_tp_rate Not yet run
unverified (harvested, no recorded run)
bit-ml/ChronoGraph/code/evaluation.py
code served (permissive licence) · get_code("142487138b00adda")
How it ran

no invocation recorded.

vae Not yet run
unverified (harvested, no recorded run)
bit-ml/ChronoGraph/code/autoencoder.py
code served (permissive licence) · get_code("ab67bb1d8ec12967")
How it ran

no invocation recorded.

vae_loss Not yet run
unverified (harvested, no recorded run)
bit-ml/ChronoGraph/code/autoencoder.py
code served (permissive licence) · get_code("253d4820e699abf6")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 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 9 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

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

Abstract

We present CHRONOGRAPH, a graph-structured multivariate time series forecasting dataset built from real-world production microservices. Each node is a service that emits a multivariate stream of system-level performance metrics, capturing CPU, memory, and network usage patterns, while directed edges encode dependencies between services. The primary task is forecasting future values of these signals at the service level. In addition, CHRONOGRAPH provides expert-annotated incident windows as anomaly labels, enabling evaluation of anomaly detection methods and assessment of forecast robustness during operational disruptions. Compared to existing benchmarks from industrial control systems or traffic and air-quality domains, CHRONOGRAPH uniquely combines (i) multivariate time series, (ii) an explicit, machine-readable dependency graph, and (iii) anomaly labels aligned with real incidents. We report baseline results spanning forecasting models, pretrained time-series foundation models, and standard anomaly detectors. CHRONOGRAPH offers a realistic benchmark for studying structure-aware forecasting and incident-aware evaluation in microservice systems. Our dataset and code are publicly available at https://github.com/bit-ml/ChronoGraph. * Equal contribution. NeurIPS 2025 Workshop on Recent Advances in Time Series Foundation Models (BERT 2 S).

For agents

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

get_harvested_code_for_paper("2509.04449")
get_code_for_paper("2509.04449")
have("2509.04449")

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

curl https://syntology.ai/api/ran/2509.04449.json

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

[![Syntology run record](https://syntology.ai/api/ran/2509.04449.svg)](https://syntology.ai/paper/2509.04449)

Connect an agent — have() is free.