SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2404.18209 · 2024

4DBInfer: A 4D Benchmarking Toolbox for Graph-Centric Predictive Modeling on Relational DBs

arXiv · PDF · Open in the Atlas

Code that ran

10 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; 1 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 8 failures of our instrument; 0 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/2404.18209.json · badge: /api/ran/2404.18209.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
awslabs/multi-table-benchmark canonical (classifier label)
classifier basis: census
10 ran · 0 no recorded run
FunctionStatusWhere it lives
dt2ts Ran
ran, honoured the contract we drafted
awslabs/multi-table-benchmark/dbinfer/datetime_utils.py
code served (permissive licence) · get_code("5f36fad7b3669524")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 1, dtype family int, shape (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 datetime64 array to pass to dt2ts
# The function expects a numpy array of datetime-like values
dt_array = np.array(['2023-01-01', '2023-01-02', '2023-01-03', '2023-01-04'], dtype='datetime64[D]')

ARGS = {
    'dt': dt_array
}
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 int, 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
infer_task_type Ran
ran; our contract draft was wrong, not the code
awslabs/multi-table-benchmark/dbinfer/evaluator.py
code served (permissive licence) · get_code("6ab61edee83a67f5")
How it ran

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

import torch
import torchmetrics.functional as MF
import torchmetrics.retrieval as MR

torch.manual_seed(0)

negated = lambda f: lambda *args, **kwargs: -f(*args, **kwargs)

def root_mean_squared_error(logits, target):
    return MF.mean_squared_error(logits, target, squared=False)

METRIC_FN = {'classification': {'accuracy': MF.accuracy, 'ap': MF.average_precision, 'auroc': MF.auroc, 'f1': MF.f1_score, 'hinge': negated(MF.hinge_loss), 'recall': MF.recall}, 'regression': {'mae': negated(MF.mean_absolute_error), 'mse': negated(MF.mean_squared_error), 'msle': negated(MF.mean_squared_log_error), 'pearson': MF.pearson_corrcoef, 'rmse': negated(root_mean_squared_error), 'r2': MF.r2_score}, 'retrieval': {'hr': MR.RetrievalHitRate(), 'mrr': MR.RetrievalMRR(), 'ndcg': MR.RetrievalNormalizedDCG()}}

def infer_task_type(metric_name: str):
    for task_type, metrics in METRIC_FN.items():
        if metric_name in metrics:
            return task_type
    raise ValueError(f'Invalid metric name {metric_name}.')

ARGS = {
    'metric_name': 'accuracy'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchmetrics — 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
load_pyd Ran
ran; our contract draft was wrong, not the code
awslabs/multi-table-benchmark/dbinfer/yaml_utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from Lilyzhangyanlin/ELF-Gym, is pointer only (licence: NONE) · get_code("1e493d91abd19387")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from Lilyzhangyanlin/ELF-Gym, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

numpy_loader Ran
ran; our contract draft was wrong, not the code
awslabs/multi-table-benchmark/dbinfer_bench/table_loader.py
code served (permissive licence) · get_code("a2c247b3e5f21c16")
How it ran

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

import numpy as np
import tempfile
import os

# Create a temporary NPZ file with sample data
temp_dir = tempfile.mkdtemp()
temp_npz_path = os.path.join(temp_dir, "test_data.npz")

# Create sample numpy arrays
array1 = np.array([[1, 2, 3], [4, 5, 6]])
array2 = np.array([7, 8, 9, 10])
array3 = np.random.randn(2, 4)

# Save to NPZ file
np.savez(temp_npz_path, data1=array1, data2=array2, data3=array3)

ARGS = {
    "path": temp_npz_path
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
parquet_loader Ran
ran; our contract draft was wrong, not the code
awslabs/multi-table-benchmark/dbinfer_bench/table_loader.py
code served (permissive licence) · get_code("2caa90eef0799c01")
How it ran

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

import pandas as pd
import numpy as np
import tempfile
import os

# Create a temporary parquet file for testing
temp_dir = tempfile.mkdtemp()
temp_path = os.path.join(temp_dir, "test.parquet")

# Create sample data and save as parquet
df = pd.DataFrame({
    'col1': np.array([1, 2, 3, 4], dtype=np.int64),
    'col2': np.array([1.5, 2.5, 3.5, 4.5], dtype=np.float64),
    'col3': np.array(['a', 'b', 'c', 'd'], dtype=object)
})
df.to_parquet(temp_path)

ARGS = {
    'path': temp_path
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
root_mean_squared_error Ran
ran; our contract draft was wrong, not the code
awslabs/multi-table-benchmark/dbinfer/evaluator.py
code served (permissive licence) · get_code("567f06c8b9cd243a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torchmetrics.functional as MF

torch.manual_seed(0)

# Create small tensors for logits and target
# batch_size=2, num_classes=4
logits = torch.randn(2, 4)
target = torch.randn(2, 4)

ARGS = {
    "logits": logits,
    "target": target,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchmetrics — 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
train_val_test_split_by_ratio Ran
ran; our contract draft was wrong, not the code
awslabs/multi-table-benchmark/dbinfer/task_construct_utils.py
code served (permissive licence) · get_code("606f7d79d85a6f5a")
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 pandas as pd
import numpy as np

# Create a simple dataframe with 100 rows and 3 columns
np.random.seed(0)
df = pd.DataFrame({
    'col1': np.random.randn(100),
    'col2': np.random.randn(100),
    'col3': np.random.randn(100)
})

ARGS = {
    'df': df,
    'train_ratio': 0.6,
    'val_ratio': 0.2
}
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
train_val_test_split_by_temporal Ran
ran; our contract draft was wrong, not the code
awslabs/multi-table-benchmark/dbinfer/task_construct_utils.py
code served (permissive licence) · get_code("070697d38b3982e5")
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
from datetime import datetime, timedelta

# Create a sample dataframe with temporal data
np.random.seed(0)
n_rows = 100
dates = [datetime(2020, 1, 1) + timedelta(days=int(i/10)) for i in range(n_rows)]
df = pd.DataFrame({
    'timestamp': dates,
    'value': np.random.randn(n_rows),
    'feature': np.random.randn(n_rows)
})

ARGS = {
    'df': df,
    'time_col': 'timestamp',
    'train_ratio': 0.6,
    'val_ratio': 0.2
}
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
train_val_test_split_shuffled Ran
ran; our contract draft was wrong, not the code
awslabs/multi-table-benchmark/dbinfer/task_construct_utils.py
code served (permissive licence) · get_code("076c1ba6bf766cc1")
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 pandas as pd
import numpy as np

# Create a small dataframe with distinct dimensions
df = pd.DataFrame({
    'col1': np.arange(100),
    'col2': np.arange(100, 200),
    'col3': np.arange(200, 300),
})

ARGS = {
    'df': df,
    'train_ratio': 0.6,
    'val_ratio': 0.2,
}
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
ts2dt Ran
ran, violated the contract we drafted
awslabs/multi-table-benchmark/dbinfer/datetime_utils.py
code served (permissive licence) · get_code("c6ebef2b8618dc9b")
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

# The function ts2dt takes a numpy array of timestamps and converts it to datetime64[ns]
# We need to provide a valid numpy array that can be converted to datetime64

# Create a simple numpy array of integers representing Unix timestamps (seconds since epoch)
ts_values = np.array([1609459200, 1609545600, 1609632000, 1609718400], dtype='int64')

ARGS = {
    'ts': ts_values
}
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

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, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 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.

FunctionAlso present inEarliest-datedOthers named
load_pyd 1 other paper
1 later
2410.12865 (2024-10)
ELF-Gym: Evaluating Large Language Models Generated Features for Tabular Prediction

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

Although RDBs store vast amounts of rich, informative data spread across interconnected tables, the progress of predictive machine learning models as applied to such tasks arguably falls well behind advances in other domains such as computer vision or natural language processing. This deficit stems, at least in part, from the lack of established/public RDB benchmarks as needed for training and evaluation purposes. As a result, related model development thus far often defaults to tabular approaches trained on ubiquitous single-table benchmarks, or on the relational side, graph-based alternatives such as GNNs applied to a completely different set of graph datasets devoid of tabular characteristics. To more precisely target RDBs lying at the nexus of these two complementary regimes, we explore a broad class of baseline models predicated on: (i) converting multi-table datasets into graphs using various strategies equipped with efficient subsampling, while preserving tabular characteristics; and (ii) trainable models with well-matched inductive biases that output predictions based on these input subgraphs. Then, to address the dearth of suitable public benchmarks and reduce siloed comparisons, we assemble a diverse collection of (i) large-scale RDB datasets and (ii) coincident predictive tasks. From a delivery standpoint, we operationalize the above four dimensions (4D) of exploration within a unified, scalable open-source toolbox called 4DBInfer. We conclude by presenting evaluations using 4DBInfer, the results of which highlight the importance of considering each such dimension in the design of RDB predictive models, as well as the limitations of more naive approaches such as simply joining adjacent tables. Our source code is released at https://github.com/awslabs/multi-table-benchmark .

For agents

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

get_harvested_code_for_paper("2404.18209")
get_code_for_paper("2404.18209")
have("2404.18209")

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

curl https://syntology.ai/api/ran/2404.18209.json

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

[![Syntology run record](https://syntology.ai/api/ran/2404.18209.svg)](https://syntology.ai/paper/2404.18209)

Connect an agent — have() is free.