SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2111.04260 · NeurIPS · 2021

Personalized Benchmarking with the Ludwig Benchmarking Toolkit

Christopher Ré, Willie Neiswanger, Karan Goel, Avanika Narayan, Piero Molino

arXiv · PDF · Open in the Atlas

Code that ran

1 of 1 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 1 has no contract verdict; 0 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/2111.04260.json · badge: /api/ran/2111.04260.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
ludwig-ai/ludwig canonical (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
FunctionStatusWhere it lives
generate_config Ran
ran on a synthesized input
ludwig-ai/ludwig/ludwig/config_generation.py
code served (permissive licence) · get_code("b1bfc6cef2090fd7")
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 json
import logging

# Set up logging
logging.basicConfig(level=logging.WARNING)
logger = logging.getLogger(__name__)

# Mock the anthropic and openai modules to avoid import errors
class MockMessage:
    def __init__(self, content):
        self.content = [MockContent(content)]

class MockContent:
    def __init__(self, text):
        self.text = text

class MockCompletion:
    def __init__(self, content):
        self.choices = [MockChoice(content)]

class MockChoice:
    def __init__(self, content):
        self.message = MockMessage(content)

# Create a valid Ludwig config as JSON string
valid_ludwig_config = {
    "input_features": [
        {"name": "age", "type": "number"},
        {"name": "income", "type": "number"},
        {"name": "purchase_history", "type": "text", "encoder": {"type": "auto_transformer"}}
    ],
    "output_features": [
        {"name": "churn", "type": "binary"},
        {"name": "lifetime_value", "type": "number"}
    ],
    "combiner": {"type": "concat"},
    "trainer": {
        "epochs": 50,
        "batch_size": 128,
        "learning_rate": 0.001
    }
}

# Mock implementation
import sys
from unittest.mock import MagicMock

# Mock anthropic module
anthropic_mock = MagicMock()
anthropic_mock.Anthropic = MagicMock(return_value=MagicMock(
    messages=MagicMock(
        create=MagicMock(return_value=MockMessage(json.dumps(valid_ludwig_config)))
    )
))
sys.modules['anthropic'] = anthropic_mock

# Mock openai module
openai_mock = MagicMock()
openai_mock.OpenAI = MagicMock(return_value=MagicMock(
    chat=MagicMock(
        completions=MagicMock(
            create=MagicMock(return_value=MockCompletion(json.dumps(valid_ludwig_config)))
        )
    )
))
sys.modules['openai'] = openai_mock

# Mock ludwig module
ludwig_mock = MagicMock()
ludwig_schema_mock = MagicMock()
ludwig_ecd_mock = MagicMock()
ludwig_base_mock = MagicMock()

class MockModelConfig:
    @staticmethod
    def from_dict(config):
        return MockModelConfig()
    
    def to_dict(self):
        return valid_ludwig_config

ludwig_base_mock.ModelConfig = MockModelConfig
ludwig_schema_mock.model_types = MagicMock()
ludwig_schema_mock.model_types.base = ludwig_base_mock
ludwig_schema_mock.model_types.ecd = ludwig_ecd_mock
ludwig_mock.schema = ludwig_schema_mock

sys.modules['ludwig'] = ludwig_mock
sys.modules['ludwig.schema'] = ludwig_schema_mock
sys.modules['ludwig.schema.model_types'] = ludwig_schema_mock.model_types
sys.modules['ludwig.schema.model_types.base'] = ludwig_base_mock
sys.modules['ludwig.schema.model_types.ecd'] = ludwig_ecd_mock

ARGS = {
    'task_description': 'I have customer data with age, income, and purchase history. I want to predict churn (binary) and lifetime value (number).',
    'model': 'claude-sonnet-4-20250514',
    'api_key': 'test-api-key-12345',
    'validate': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
anthropic, ludwig, openai — 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

Also present in other papers' code

Of 1 samples lifted from this paper's linked repositories, 1 carry a function identity and 0 do not. Of the 1 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 1 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

The rapid proliferation of machine learning models across domains and deployment settings has given rise to various communities (e.g. industry practitioners) which seek to benchmark models across tasks and objectives of personal value. Unfortunately, these users cannot use standard benchmark results to perform such value-driven comparisons as traditional benchmarks evaluate models on a single objective (e.g. average accuracy) and fail to facilitate a standardized training framework that controls for confounding variables (e.g. computational budget), making fair comparisons difficult. To address these challenges, we introduce the open-source Ludwig Benchmarking Toolkit (LBT), a personalized benchmarking toolkit for running end-to-end benchmark studies (from hyperparameter optimization to evaluation) across an easily extensible set of tasks, deep learning models, datasets and evaluation metrics. LBT provides a configurable interface for controlling training and customizing evaluation, a standardized training framework for eliminating confounding variables, and support for multi-objective evaluation. We demonstrate how LBT can be used to create personalized benchmark studies with a large-scale comparative analysis for text classification across 7 models and 9 datasets. We explore the trade-offs between inference latency and performance, relationships between dataset attributes and performance, and the effects of pretraining on convergence and robustness, showing how LBT can be used to satisfy various benchmarking objectives.

For agents

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

get_harvested_code_for_paper("2111.04260")
get_code_for_paper("2111.04260")
have("2111.04260")

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

curl https://syntology.ai/api/ran/2111.04260.json

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

[![Syntology run record](https://syntology.ai/api/ran/2111.04260.svg)](https://syntology.ai/paper/2111.04260)

Connect an agent — have() is free.