SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1806.02382 · 2018

Variational Autoencoder with Arbitrary Conditioning

arXiv · PDF · Open in the Atlas

Code that ran

5 of 7 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 5 have no contract verdict; 0 failures of our instrument; 2 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/1806.02382.json · badge: /api/ran/1806.02382.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
tigvarts/ucm canonical (classifier label)
classifier basis: census
5 ran · 2 no recorded run
FunctionStatusWhere it lives
categorical_parse_params_column Ran
ran on a synthesized input
tigvarts/ucm/prob_utils.py
code served (permissive licence) · get_code("ce53b3b51ff21c74")
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 Categorical. 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.distributions import Categorical
from torch.nn.functional import softmax

torch.manual_seed(0)

# Create a 2D tensor with shape (batch_size=2, num_classes=4)
# This represents logits for a batch of 2 samples with 4 categorical classes
params = torch.randn(2, 4)

ARGS = {
    "params": params,
    "min_prob": 0,
}
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 Categorical — a Categorical output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Categorical 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
extend_batch_tuple Ran
ran on a synthesized input
tigvarts/ucm/train_utils.py
code served (permissive licence) · get_code("629c1fe4bfc34dc4")
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. 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.utils.data import DataLoader, TensorDataset

torch.manual_seed(0)

# Create small sample tensors for a dataset
x_data = torch.randn(5, 4)  # 5 samples, 4 features
y_data = torch.randn(5, 3)  # 5 samples, 3 targets

# Create a dataset and dataloader
dataset = TensorDataset(x_data, y_data)
dataloader = DataLoader(dataset, batch_size=2, shuffle=False)

# Create initial batch (list of tensors with batch_size=2)
initial_batch = [torch.randn(2, 4), torch.randn(2, 3)]

# Target batch size to extend to
target_batch_size = 4

ARGS = {
    "batch": initial_batch,
    "dataloader": dataloader,
    "batch_size": target_batch_size
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list — a list 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
get_validation_iwae Ran
ran on a synthesized input
tigvarts/ucm/train_utils.py
code served (permissive licence) · get_code("ce09de21e4235012")
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. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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)

# Create a minimal VAEAC-like model that implements batch_iwae
class MinimalVAEAC(nn.Module):
    def __init__(self, data_dim, latent_dim):
        super().__init__()
        self.data_dim = data_dim
        self.latent_dim = latent_dim
        self.encoder = nn.Linear(data_dim, latent_dim)
        self.decoder = nn.Linear(latent_dim, data_dim)
    
    def batch_iwae(self, batch, mask, num_samples):
        """
        Compute IWAE log likelihood.
        batch: (batch_size, data_dim)
        mask: (batch_size, data_dim)
        num_samples: int
        Returns: (batch_size,) tensor of IWAE estimates
        """
        batch_size = batch.shape[0]
        # Simple mock IWAE computation - just return random values per sample
        return torch.randn(batch_size)

# Create validation dataloader (list of tensors)
batch_size = 4
data_dim = 3
num_batches = 2
val_dataloader = [torch.randn(2, data_dim) for _ in range(num_batches)]

# Create mask generator function
def mask_generator(batch):
    """Generate random binary masks of same shape as batch"""
    return torch.bernoulli(torch.ones_like(batch) * 0.5)

# Create model
model = MinimalVAEAC(data_dim=data_dim, latent_dim=2)

# Create arguments dict
ARGS = {
    'val_dataloader': val_dataloader,
    'mask_generator': mask_generator,
    'batch_size': batch_size,
    'model': model,
    'num_samples': 3,
    'verbose': False
}
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 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
normal_parse_params Ran
ran on a synthesized input
tigvarts/ucm/prob_utils.py
code served (permissive licence) · get_code("8cf9a01573b92861")
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 Normal. 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.distributions import Normal
from torch.nn.functional import softplus

torch.manual_seed(0)

# Create a small tensor with batch size 2 and dimension 4
# The function expects params with shape (batch_size, 2*feature_dim)
# so we need an even number of channels
params = torch.randn(2, 4)

min_sigma = 0.1

ARGS = {
    "params": params,
    "min_sigma": min_sigma,
}
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 Normal — a Normal output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Normal 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
optimizer Ran
ran on a synthesized input
tigvarts/ucm/celeba_model/model.py
code served (permissive licence) · get_code("bb98b06acff24a8c")
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 Adam. 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
from torch.optim import Adam

torch.manual_seed(0)

# Create a simple neural network with parameters
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()

ARGS = {
    "parameters": model.parameters()
}
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 Adam — a Adam output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Adam 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
compute_normalization Not yet run
unverified (harvested, no recorded run)
tigvarts/ucm/datasets.py
code served (permissive licence) · get_code("f195c200bfc3a313")
How it ran

no invocation recorded.

extend_batch Not yet run
unverified (harvested, no recorded run)
tigvarts/ucm/train_utils.py
code served (permissive licence) · get_code("6db2d0c299759c56")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 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 7 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 propose a single neural probabilistic model based on variational autoencoder that can be conditioned on an arbitrary subset of observed features and then sample the remaining features in "one shot". The features may be both real-valued and categorical. Training of the model is performed by stochastic variational Bayes. The experimental evaluation on synthetic data, as well as feature imputation and image inpainting problems, shows the effectiveness of the proposed approach and diversity of the generated samples.

For agents

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

get_harvested_code_for_paper("1806.02382")
get_code_for_paper("1806.02382")
have("1806.02382")

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

curl https://syntology.ai/api/ran/1806.02382.json

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

[![Syntology run record](https://syntology.ai/api/ran/1806.02382.svg)](https://syntology.ai/paper/1806.02382)

Connect an agent — have() is free.