SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2209.15421 · 2022

TabDDPM: Modelling Tabular Data with Diffusion Models

arXiv · PDF · Open in the Atlas

Code that ran

22 of 22 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; 4 honoured the contract we drafted, 1 violated it and 5 have no contract verdict; 12 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/2209.15421.json · badge: /api/ran/2209.15421.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
rotot0/tab-ddpm canonical (classifier label)
classifier basis not recorded on the link
16 ran · 0 no recorded run
yifei-liu-stat/syn community (classifier label)
classifier basis not recorded on the link
6 ran · 0 no recorded run
FunctionStatusWhere it lives
approx_standard_normal_cdf Ran
ran, honoured the contract we drafted
rotot0/tab-ddpm/tab_ddpm/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from tsinghua-fib-lab/Spatio-temporal-Diffusion-Point-Processes, is pointer only (licence: NONE) · get_code("9deab2a47adbf3b7")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from tsinghua-fib-lab/Spatio-temporal-Diffusion-Point-Processes, 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.

betas_for_alpha_bar Ran
ran, honoured the contract we drafted
rotot0/tab-ddpm/tab_ddpm/gaussian_multinomial_diffsuion.py
pointer only (licence: MIT); the stored copy get_code serves, from blt2114/twisted_diffusion_sampler, is pointer only (licence: NONE) · get_code("6cdc8772f5db6f48")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from blt2114/twisted_diffusion_sampler, 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.

calculate_rmse Ran
ran, violated the contract we drafted
rotot0/tab-ddpm/lib/metrics.py
code served (permissive licence) · get_code("f0c2655e22e2ed2d")
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 float, 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

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

# Create small sample arrays for y_true and y_pred
y_true = np.array([1.0, 2.0, 3.0, 4.0, 5.0])
y_pred = np.array([1.1, 2.2, 2.9, 4.1, 4.8])

# Create std parameter (optional float)
std = 2.5

# Construct the arguments dictionary
ARGS = {
    "y_true": y_true,
    "y_pred": y_pred,
    "std": std
}
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 float, 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
concat_data Ran
ran on a synthesized input
yifei-liu-stat/syn/tab-ddpm/utils/utils_syn.py
code served (permissive licence) · get_code("52f84e39cc40d193")
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 os
import numpy as np
import pandas as pd
import tempfile
import shutil

# Create a temporary directory for test data
temp_dir = tempfile.mkdtemp()

# Create sample data files
# y_train.npy - response variable (1D array)
y_train = np.array([0, 1, 0, 1])
np.save(os.path.join(temp_dir, 'y_train.npy'), y_train)

# X_num_train.npy - numerical features (4 samples, 3 numerical features)
X_num_train = np.array([
    [1.5, 2.3, 3.1],
    [2.1, 3.2, 4.5],
    [0.9, 1.8, 2.7],
    [3.2, 4.1, 5.6]
])
np.save(os.path.join(temp_dir, 'X_num_train.npy'), X_num_train)

# X_cat_train.npy - categorical features (4 samples, 2 categorical features)
X_cat_train = np.array([
    ['a', 'x'],
    ['b', 'y'],
    ['a', 'z'],
    ['c', 'x']
], dtype=object)
np.save(os.path.join(temp_dir, 'X_cat_train.npy'), X_cat_train)

ARGS = {
    'data_path': temp_dir,
    'split': 'train',
    'num_features_list': ['num_feature_1', 'num_feature_2', 'num_feature_3'],
    'cat_features_list': ['cat_feature_1', 'cat_feature_2'],
    'y_feature': 'target',
    'is_y_cat': False
}
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 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
cos_sin Ran
ran; our contract draft was wrong, not the code
rotot0/tab-ddpm/lib/deep.py
code served (permissive licence) · get_code("61b240fd493b6733")
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, 8) 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
from torch import Tensor

torch.manual_seed(0)

# The function takes a single parameter x which is a Tensor
# It applies torch.cos and torch.sin to it, then concatenates along the last dimension
# So x should be a tensor of any shape

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, shape (2, 8) — shape (2, 8) 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
discretized_gaussian_log_likelihood Ran
ran; our contract draft was wrong, not the code
rotot0/tab-ddpm/tab_ddpm/utils.py
code served (permissive licence) · get_code("6b6f39a3cb3edfa0")
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, 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 torch
import numpy as np

torch.manual_seed(0)

# Create small tensors with distinct sizes for different dimensions
batch_size = 2
dim = 4

# x: target images rescaled to [-1, 1]
x = torch.randn(batch_size, dim) * 0.5

# means: Gaussian mean, same shape as x
means = torch.randn(batch_size, dim) * 0.3

# log_scales: Gaussian log stddev, same shape as x
log_scales = torch.randn(batch_size, dim) * 0.1

ARGS = {
    "x": x,
    "means": means,
    "log_scales": log_scales,
}
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, 4) — shape (2, 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
geglu Ran
ran; our contract draft was wrong, not the code
rotot0/tab-ddpm/tab_ddpm/modules.py
pointer only (licence: MIT); the stored copy get_code serves, from fangzy96/tabcutmix, is pointer only (licence: NONE) · get_code("49f8fe0655f00ee5")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from fangzy96/tabcutmix, 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.

generate_imputation Ran
ran on a synthesized input
yifei-liu-stat/syn/conditional/utils/ddpm.py
code served (permissive licence) · get_code("f1fa7c754e7515d6")
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, 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 torch
import torch.nn as nn

torch.manual_seed(0)

class MockDDPM(nn.Module):
    def __init__(self):
        super().__init__()
        self.device = torch.device('cpu')
        self.n_steps = 5
        
        # Create alphas, alpha_bars, and betas
        self.register_buffer('alphas', torch.linspace(0.99, 0.90, self.n_steps))
        self.register_buffer('alpha_bars', torch.linspace(0.99, 0.50, self.n_steps))
        self.register_buffer('betas', torch.linspace(0.01, 0.10, self.n_steps))
        
        # Simple linear layer for forward pass
        self.fc = nn.Linear(4, 4)
    
    def forward(self, x, time_tensor):
        """Called as ddpm(x0, time_tensor) to get x_known"""
        return self.fc(x)
    
    def backward(self, x, time_tensor):
        """Called as ddpm.backward(x, time_tensor) to get eta_theta"""
        return self.fc(x)

# Create mock DDPM model
ddpm = MockDDPM()

# Create input tensors
batch_size = 2
feature_dim = 4
yx = torch.randn(batch_size, feature_dim)
mask_yx = torch.bernoulli(torch.full((batch_size, feature_dim), 0.7))

ARGS = {
    'ddpm': ddpm,
    'yx': yx,
    'mask_yx': mask_yx,
    'resampling_steps': 1,
    'device': None,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, tqdm — 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, 4) — shape (2, 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
generate_samples Ran
ran on a synthesized input
yifei-liu-stat/syn/conditional/utils/ddpm.py
code served (permissive licence) · get_code("0ba9a28441c16b29")
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, 8) 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 torch.nn as nn

torch.manual_seed(0)

class SimpleDDPM(nn.Module):
    def __init__(self, tabular_dim=8, n_steps=10):
        super().__init__()
        self.tabular_dim = tabular_dim
        self.n_steps = n_steps
        self.device = torch.device('cpu')
        
        # Create schedules for alphas and betas
        betas = torch.linspace(0.0001, 0.02, n_steps)
        self.register_buffer('betas', betas)
        
        alphas = 1 - betas
        self.register_buffer('alphas', alphas)
        
        alpha_bars = torch.cumprod(alphas, dim=0)
        self.register_buffer('alpha_bars', alpha_bars)
        
        # Simple neural network for backward pass
        self.net = nn.Sequential(
            nn.Linear(tabular_dim + 1, 16),
            nn.ReLU(),
            nn.Linear(16, tabular_dim)
        )
    
    def backward(self, x, time_tensor):
        """Returns predicted noise eta_theta given x and t"""
        # Embed time as a single scalar feature
        batch_size = x.shape[0]
        time_embed = time_tensor.float().unsqueeze(1) / self.n_steps
        x_with_time = torch.cat([x, time_embed], dim=1)
        eta_theta = self.net(x_with_time)
        return eta_theta

ddpm = SimpleDDPM(tabular_dim=8, n_steps=10)

ARGS = {
    'ddpm': ddpm,
    'n_samples': 2,
    'device': None,
    'tabular_dim': 8
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, tqdm — 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, 8) — shape (2, 8) 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_category_sizes Ran
ran, honoured the contract we drafted
rotot0/tab-ddpm/lib/data.py
code served (permissive licence) · get_code("4d744cd924d61d6f")
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 int; 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
import numpy as np

torch.manual_seed(0)

# The function get_category_sizes takes X which can be either a torch.Tensor or np.ndarray
# It transposes X and converts to a list, then counts unique values in each row
# We need to create test data with categorical values

# Create a torch.Tensor with categorical values (batch_size=2, num_features=4)
X_torch = torch.tensor([
    [0, 1, 0, 2],
    [1, 1, 2, 0]
], dtype=torch.long)

# Create an np.ndarray with categorical values (batch_size=2, num_features=4)
X_numpy = np.array([
    [0, 1, 0, 2],
    [1, 1, 2, 0]
], dtype=np.int32)

# Test with torch.Tensor
ARGS = {
    "X": X_torch
}
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 list, ndim 1, dtype family int — 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
get_n_parameters Ran
ran, honoured the contract we drafted
rotot0/tab-ddpm/lib/deep.py
code served (permissive licence) · get_code("4bb318a745ffe343")
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 int, ndim 0, dtype family int, 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 torch.nn as nn

torch.manual_seed(0)

# Build a small nn.Module for testing
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

model = SimpleModel()

# Construct ARGS dictionary
ARGS = {
    "m": model
}
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 int, ndim 0, dtype family int, 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
get_named_beta_schedule Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
rotot0/tab-ddpm/tab_ddpm/gaussian_multinomial_diffsuion.py
code served (permissive licence) · get_code("141937afc98f8d1e")
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) and returned ndarray, ndim 1, dtype family float, shape (100,) 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 numpy as np

# Test with both schedule types
ARGS = {
    'schedule_name': 'cosine',
    'num_diffusion_timesteps': 100
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 (100,) — shape (100,) 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_path Ran
ran; our contract draft was wrong, not the code
rotot0/tab-ddpm/lib/env.py
code served (permissive licence) · get_code("eac7b341abfe7944")
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 PosixPath. 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 typing as ty
from pathlib import Path
import tempfile
import os

PROJ = Path('tab-ddpm/').absolute().resolve()

def get_path(path: ty.Union[str, Path]) -> Path:
    if isinstance(path, str):
        path = Path(path)
    if not path.is_absolute():
        path = PROJ / path
    return path.resolve()

# Create a temporary directory to work with
temp_dir = tempfile.mkdtemp()

# Test with a relative path string
relative_path_str = "data/dataset.csv"

# Test with a relative path Path object
relative_path_obj = Path("models/checkpoint.pt")

# Test with an absolute path string
absolute_path_str = os.path.join(temp_dir, "absolute_file.txt")

# Test with an absolute path Path object
absolute_path_obj = Path(temp_dir) / "another_file.pt"

# ARGS dict with a relative string path as the primary test case
ARGS = {
    "path": relative_path_str
}
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 PosixPath — a PosixPath output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a PosixPath 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
get_relative_path Ran
ran; our contract draft was wrong, not the code
rotot0/tab-ddpm/lib/env.py
code served (permissive licence) · get_code("170e57097113c6f4")
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 PosixPath. 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 typing as ty
from pathlib import Path

PROJ = Path('tab-ddpm/').absolute().resolve()

def get_path(path: ty.Union[str, Path]) -> Path:
    if isinstance(path, str):
        path = Path(path)
    if not path.is_absolute():
        path = PROJ / path
    return path.resolve()

def get_relative_path(path: ty.Union[str, Path]) -> Path:
    return get_path(path).relative_to(PROJ)

# Setup code
ARGS = {
    "path": "data/train.csv"
}
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 PosixPath — a PosixPath output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a PosixPath 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
imdb_jsonl_to_np Ran
ran on a synthesized input
yifei-liu-stat/syn/sentiment/imdb_distillbert.py
code served (permissive licence) · get_code("c48a02ede7b161d4")
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, 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 numpy as np
import json
import tempfile
import os

# Create a temporary JSONL file with valid IMDB-format data
temp_file = tempfile.NamedTemporaryFile(mode='w', suffix='.jsonl', delete=False)
temp_file_path = temp_file.name

# Write sample JSONL entries matching the function's expected format
sample_entries = [
    {'prompt': 'This movie was great\n\n###\n\n', 'completion': ' 1'},
    {'prompt': 'This movie was terrible\n\n###\n\n', 'completion': ' 0'},
    {'prompt': 'Amazing film\n\n###\n\n', 'completion': ' 1'},
]

for entry in sample_entries:
    temp_file.write(json.dumps(entry) + '\n')

temp_file.close()

ARGS = {
    'jsonl_data_path': temp_file_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 tuple, ndim 2, dtype family float — a tuple 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
normal_kl Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
rotot0/tab-ddpm/tab_ddpm/utils.py
code served (permissive licence) · get_code("4707efa98ac6b989")
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, 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 torch

torch.manual_seed(0)

# Create small tensors for KL divergence computation
# The function expects mean and logvar parameters for two Gaussians
mean1 = torch.randn(2, 4)
logvar1 = torch.randn(2, 4)
mean2 = torch.randn(2, 4)
logvar2 = torch.randn(2, 4)

ARGS = {
    "mean1": mean1,
    "logvar1": logvar1,
    "mean2": mean2,
    "logvar2": logvar2,
}
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, shape (2, 4) — shape (2, 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
pack_config Ran
ran; our contract draft was wrong, not the code
rotot0/tab-ddpm/lib/util.py
code served (permissive licence) · get_code("82071b4852fbdb59")
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 torch
import numpy as np
from typing import Any, Dict

torch.manual_seed(0)

# The function pack_config takes a RawConfig (Dict[str, Any]) parameter
# It recursively traverses the config dict/list structure and replaces None values
# with the string '__none__'

# Create a sample config with various nested structures including None values
config = {
    "model_name": "test_model",
    "learning_rate": 0.001,
    "batch_size": 32,
    "dropout": None,
    "layers": [1, 2, None, 4],
    "nested": {
        "param1": "value1",
        "param2": None,
        "deep": {
            "value": None,
            "other": 42
        }
    },
    "mixed_list": [None, "text", {"key": None}],
}

ARGS = {
    "config": config
}
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 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
prepare_dl Ran
ran on a synthesized input
yifei-liu-stat/syn/sentiment/imdb_distillbert.py
code served (permissive licence) · get_code("d9a0fc20c06a8d77")
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 DataLoader. 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
from torch import nn
from torch.utils.data import Dataset, DataLoader

torch.manual_seed(0)

# Create a minimal tokenizer stand-in that has encode_plus method
class MinimalTokenizer:
    def encode_plus(self, text, add_special_tokens=True, max_length=512, 
                   return_token_type_ids=False, padding='max_length', 
                   truncation=True, return_attention_mask=True, return_tensors='pt'):
        # Return minimal valid encoding
        input_ids = torch.ones(1, max_length, dtype=torch.long)
        attention_mask = torch.ones(1, max_length, dtype=torch.long)
        return {
            'input_ids': input_ids,
            'attention_mask': attention_mask
        }

# Create minimal test data
reviews = [
    "This movie was great!",
    "I did not like this film."
]

sentiments = [1, 0]

tokenizer = MinimalTokenizer()

ARGS = {
    'reviews': reviews,
    'sentiments': sentiments,
    'tokenizer': tokenizer
}
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 DataLoader — a DataLoader output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataLoader 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
reglu Ran
ran; our contract draft was wrong, not the code
rotot0/tab-ddpm/tab_ddpm/modules.py
pointer only (licence: MIT); the stored copy get_code serves, from fangzy96/tabcutmix, is pointer only (licence: NONE) · get_code("58bd3831c7fb6729")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from fangzy96/tabcutmix, 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.

timestep_embedding Ran
ran; our contract draft was wrong, not the code
rotot0/tab-ddpm/tab_ddpm/modules.py
code served (permissive licence) · get_code("fe2398f0b08da143")
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 (4, 8) 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 math

torch.manual_seed(0)

# Create valid arguments for timestep_embedding function
timesteps = torch.tensor([0, 1, 2, 3], dtype=torch.float32)  # 1-D tensor of batch size 4
dim = 8  # output dimension
max_period = 10000  # default max_period

ARGS = {
    "timesteps": timesteps,
    "dim": dim,
    "max_period": max_period,
}
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, shape (4, 8) — shape (4, 8) 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
timestep_embedding Ran
ran; our contract draft was wrong, not the code
yifei-liu-stat/syn/conditional/utils/utils_model.py
code served (permissive licence) · get_code("80360f1cea3fe1f1")
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, 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 torch
import math

torch.manual_seed(0)

# The function takes:
# - timesteps: 1-D Tensor of N indices
# - dim: integer dimension
# - max_period: integer controlling frequency (has default)

# Create a small batch of timestep indices
timesteps = torch.tensor([0, 5], dtype=torch.int64)

# Dimension for the embedding output
dim = 4

# Max period parameter (using default is fine, but we can also specify it)
max_period = 10000

ARGS = {
    "timesteps": timesteps,
    "dim": dim,
    "max_period": max_period,
}
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, shape (2, 4) — shape (2, 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
unpack_config Ran
ran; our contract draft was wrong, not the code
rotot0/tab-ddpm/lib/util.py
code served (permissive licence) · get_code("3a4a181a3fa716f2")
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 torch
import numpy as np
from typing import Any, Dict

torch.manual_seed(0)

# The function unpack_config takes a RawConfig (Dict[str, Any])
# and replaces all instances of the string '__none__' with None
# It uses the _replace helper which recursively traverses dicts and lists

# Create a config with various data types including the sentinel value
config_with_none_sentinel = {
    "model_name": "bert",
    "hidden_size": 768,
    "num_layers": 12,
    "dropout": "__none__",
    "activation": "gelu",
    "nested": {
        "key1": "value1",
        "key2": "__none__",
        "deep": {
            "param": "__none__",
            "value": 42
        }
    },
    "list_data": [1, 2, "__none__", 4],
    "mixed_list": [
        {"a": "__none__", "b": 10},
        "__none__",
        "keep_this"
    ]
}

ARGS = {
    "config": config_with_none_sentinel
}
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 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 22 samples lifted from this paper's linked repositories, 22 carry a function identity and 0 do not. Of the 21 distinct functions with an identity, 16 are also present in other papers whose code we hold (13 with at least one paper dated earlier, 1 the same month, 16 later) and 5 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
betas_for_alpha_bar 92 other papers
18 dated earlier, 74 later · 3 only as an assembly brick
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2010.02502 (2020-10)
Denoising Diffusion Implicit Models
2102.09672 (2021-02)
Improved Denoising Diffusion Probabilistic Models
and 89 more not named here
get_named_beta_schedule 63 other papers
14 dated earlier, 49 later
2010.02502 (2020-10)
Denoising Diffusion Implicit Models
2108.02938 (2021-08)
ILVR: Conditioning Method for Denoising Diffusion Probabilistic Models
2111.11672 (2021-11)
Few-shot Image Generation with Mixup-based Distance Learning
and 60 more not named here
normal_kl 34 other papers
1 dated earlier, 33 later
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2211.13221 (2022-11)
Latent Video Diffusion Models for High-Fidelity Long Video Generation
and 31 more not named here
timestep_embedding 29 other papers
3 dated earlier, 1 the same month, 25 later · 9 only as an assembly brick
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2010.11929 (2020-10)
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
2103.11161 (2021-03)
MonteFloor: Extending MCTS for Reconstructing Accurate Large-Scale Floor Plans
and 26 more not named here
approx_standard_normal_cdf 10 other papers
1 dated earlier, 9 later
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2303.18242 (2023-03)
∞-Diff: Infinite Resolution Diffusion with Subsampled Mollified States
and 7 more not named here
discretized_gaussian_log_likelihood 8 other papers
1 dated earlier, 7 later
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2310.03502 (2023-10)
Kandinsky: an Improved Text-to-Image Synthesis with Image Prior and Latent Diffusion
and 5 more not named here
geglu 8 other papers
2 dated earlier, 6 later
2102.08921 (2021-02)
How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Models
2207.03208 (2022-07)
Revisiting Pretraining Objectives for Tabular Deep Learning
2309.16220 (2023-09)
Unmasking the Chameleons: A Benchmark for Out-of-Distribution Detection in Medical Tabular Data
and 5 more not named here
reglu 8 other papers
2 dated earlier, 6 later
2102.08921 (2021-02)
How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Models
2207.03208 (2022-07)
Revisiting Pretraining Objectives for Tabular Deep Learning
2309.16220 (2023-09)
Unmasking the Chameleons: A Benchmark for Out-of-Distribution Detection in Medical Tabular Data
and 5 more not named here
cos_sin 3 other papers
1 dated earlier, 2 later
2203.05556 (2022-03)
On Embeddings for Numerical Features in Tabular Deep Learning
Tiong_Flexible_Biometrics_Recognition_Bridging_the_Multimodality_Gap_through_Attention_Alignment_CVPR_2024_paper (2024 (year only))
Flexible Biometrics Recognition: Bridging the Multimodality Gap through Attention, Alignment and Prompt Tuning
2601.22816 (2026-01)
Cascaded Flow Matching for Heterogeneous Tabular Data with Mixed-Type Features
get_category_sizes 3 other papers
3 later
2211.16887 (2022-11)
T2G-Former: Organizing Tabular Features into Relation Graphs Promotes Heterogeneous Feature Interaction
2410.20626 (2024-10)
TabDiff: a Mixed-type Diffusion Model for Tabular Data Generation
2601.22816 (2026-01)
Cascaded Flow Matching for Heterogeneous Tabular Data with Mixed-Type Features
get_n_parameters 3 other papers
1 dated earlier, 2 later
2203.05556 (2022-03)
On Embeddings for Numerical Features in Tabular Deep Learning
2406.19380 (2024-06)
TabReD: Analyzing Pitfalls and Filling the Gaps in Tabular Deep Learning Benchmarks
2601.22816 (2026-01)
Cascaded Flow Matching for Heterogeneous Tabular Data with Mixed-Type Features
get_path 3 other papers
3 later
2211.16887 (2022-11)
T2G-Former: Organizing Tabular Features into Relation Graphs Promotes Heterogeneous Feature Interaction
2410.20626 (2024-10)
TabDiff: a Mixed-type Diffusion Model for Tabular Data Generation
2601.22816 (2026-01)
Cascaded Flow Matching for Heterogeneous Tabular Data with Mixed-Type Features
get_relative_path 3 other papers
3 later
2211.16887 (2022-11)
T2G-Former: Organizing Tabular Features into Relation Graphs Promotes Heterogeneous Feature Interaction
2410.20626 (2024-10)
TabDiff: a Mixed-type Diffusion Model for Tabular Data Generation
2601.22816 (2026-01)
Cascaded Flow Matching for Heterogeneous Tabular Data with Mixed-Type Features
calculate_rmse 2 other papers
1 dated earlier, 1 later
2203.05556 (2022-03)
On Embeddings for Numerical Features in Tabular Deep Learning
2601.22816 (2026-01)
Cascaded Flow Matching for Heterogeneous Tabular Data with Mixed-Type Features
pack_config 2 other papers
1 dated earlier, 1 later
2203.05556 (2022-03)
On Embeddings for Numerical Features in Tabular Deep Learning
2601.22816 (2026-01)
Cascaded Flow Matching for Heterogeneous Tabular Data with Mixed-Type Features
unpack_config 2 other papers
1 dated earlier, 1 later
2203.05556 (2022-03)
On Embeddings for Numerical Features in Tabular Deep Learning
2601.22816 (2026-01)
Cascaded Flow Matching for Heterogeneous Tabular Data with Mixed-Type Features

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

Denoising diffusion probabilistic models are currently becoming the leading paradigm of generative modeling for many important data modalities. Being the most prevalent in the computer vision community, diffusion models have also recently gained some attention in other domains, including speech, NLP, and graph-like data. In this work, we investigate if the framework of diffusion models can be advantageous for general tabular problems, where datapoints are typically represented by vectors of heterogeneous features. The inherent heterogeneity of tabular data makes it quite challenging for accurate modeling, since the individual features can be of completely different nature, i.e., some of them can be continuous and some of them can be discrete. To address such data types, we introduce TabDDPM -- a diffusion model that can be universally applied to any tabular dataset and handles any type of feature. We extensively evaluate TabDDPM on a wide set of benchmarks and demonstrate its superiority over existing GAN/VAE alternatives, which is consistent with the advantage of diffusion models in other fields. Additionally, we show that TabDDPM is eligible for privacy-oriented setups, where the original datapoints cannot be publicly shared.

For agents

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

get_harvested_code_for_paper("2209.15421")
get_code_for_paper("2209.15421")
have("2209.15421")

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

curl https://syntology.ai/api/ran/2209.15421.json

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

[![Syntology run record](https://syntology.ai/api/ran/2209.15421.svg)](https://syntology.ai/paper/2209.15421)

Connect an agent — have() is free.