SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2603.26746 · 2026

TDEC: Deep Embedded Image Clustering with Transformer and Distribution Information

Haiyang Zheng, Hongpeng Wang, Ruilin Zhang

arXiv · PDF · Open in the Atlas

Code that ran

16 of 23 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 1 violated it and 4 have no contract verdict; 11 failures of our instrument; 7 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2603.26746.json · badge: /api/ran/2603.26746.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
BGU-CS-VIL/DeepDPM canonical (classifier label)
classifier basis: census
7 ran · 1 no recorded run
XifengGuo/DEC-keras canonical (classifier label)
classifier basis: census
3 ran · 1 no recorded run
XifengGuo/ASPC-DA community
Papers with Code archive
3 ran · 3 no recorded run
JinyuCai95/EDESC-pytorch community
Papers with Code archive
3 ran · 2 no recorded run
FunctionStatusWhere it lives
AE Ran
ran on a synthesized input
JinyuCai95/EDESC-pytorch/EDESC.py
code served (permissive licence) · get_code("fa8b00d7b7812f8a")
How it ran

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

import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.nn import Linear

class AE(nn.Module):

    def __init__(self, n_enc_1, n_enc_2, n_enc_3, n_dec_1, n_dec_2, n_dec_3, n_input, n_z):
        super(AE, self).__init__()
        self.enc_1 = Linear(n_input, n_enc_1)
        self.enc_2 = Linear(n_enc_1, n_enc_2)
        self.enc_3 = Linear(n_enc_2, n_enc_3)
        self.z_layer = Linear(n_enc_3, n_z)
        self.dec_1 = Linear(n_z, n_dec_1)
        self.dec_2 = Linear(n_dec_1, n_dec_2)
        self.dec_3 = Linear(n_dec_2, n_dec_3)
        self.x_bar_layer = Linear(n_dec_3, n_input)

    def forward(self, x):
        enc_h1 = F.relu(self.enc_1(x))
        enc_h2 = F.relu(self.enc_2(enc_h1))
        enc_h3 = F.relu(self.enc_3(enc_h2))
        z = self.z_layer(enc_h3)
        dec_h1 = F.relu(self.dec_1(z))
        dec_h2 = F.relu(self.dec_2(dec_h1))
        dec_h3 = F.relu(self.dec_3(dec_h2))
        x_bar = self.x_bar_layer(dec_h3)
        return (x_bar, z)

INIT_ARGS = {
    'n_input': 4,
    'n_enc_1': 8,
    'n_enc_2': 6,
    'n_enc_3': 5,
    'n_z': 3,
    'n_dec_1': 5,
    'n_dec_2': 6,
    'n_dec_3': 8,
}

ARGS = {
    'x': torch.randn(2, 4),
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
D_constraint1 Ran
ran on a synthesized input
JinyuCai95/EDESC-pytorch/EDESC.py
code served (permissive licence) · get_code("dd7198b7ad25b6ed")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 0, dtype family float; no output shape is stated. 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 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

class D_constraint1(torch.nn.Module):

    def __init__(self):
        super(D_constraint1, self).__init__()

    def forward(self, d):
        I = torch.eye(d.shape[1]).cuda()
        loss_d1_constraint = torch.norm(torch.mm(d.t(), d) * I - I)
        return 0.001 * loss_d1_constraint


INIT_ARGS = {}

ARGS = {
    'd': torch.randn(4, 3)
}

if __name__ == '__main__':
    model = D_constraint1(**INIT_ARGS)
    try:
        output = model(**ARGS)
        print(f"Output: {output}")
    except RuntimeError as e:
        print(f"Expected error (CUDA not available): {e}")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 0, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
D_constraint2 Ran
ran on a synthesized input
JinyuCai95/EDESC-pytorch/EDESC.py
code served (permissive licence) · get_code("530995e332e47288")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 0, dtype family float; no output shape is stated. 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 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

class D_constraint2(torch.nn.Module):

    def __init__(self):
        super(D_constraint2, self).__init__()

    def forward(self, d, dim, n_clusters):
        S = torch.ones(d.shape[1], d.shape[1]).cuda()
        zero = torch.zeros(dim, dim)
        for i in range(n_clusters):
            S[i * dim:(i + 1) * dim, i * dim:(i + 1) * dim] = zero
        loss_d2_constraint = torch.norm(torch.mm(d.t(), d) * S)
        return 0.001 * loss_d2_constraint


INIT_ARGS = {}

ARGS = {
    'd': torch.randn(5, 8),
    'dim': 4,
    'n_clusters': 2,
}

if __name__ == '__main__':
    model = D_constraint2(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Output value: {output}")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 0, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
MyIterator Ran
ran on a synthesized input · constructed an object; no result was computed
XifengGuo/ASPC-DA/ASPC.py
pointer only (licence: NONE) · get_code("49cb04d739f2a5ec")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, XifengGuo/ASPC-DA 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.

autoencoder Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
XifengGuo/ASPC-DA/ASPC.py
pointer only (licence: NONE) · get_code("29ce8cbdbf6cdafc")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, XifengGuo/ASPC-DA 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.

autoencoder Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
XifengGuo/DEC-keras/DEC.py
code served (permissive licence) · get_code("784029c1f5854992")
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 1, 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

ARGS = {
    'dims': [10, 8, 6, 4],
    'act': 'relu',
    'init': 'glorot_uniform'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
keras, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, 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
best_cluster_fit Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
BGU-CS-VIL/DeepDPM/src/utils.py
code served (permissive licence) · get_code("2daf9e1c80f03499")
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

# Construct small numpy arrays for y_true and y_pred
# These should be 1D arrays of cluster labels
y_true = np.array([0, 0, 1, 1, 2, 2], dtype=np.int64)
y_pred = np.array([0, 0, 1, 2, 2, 1], dtype=np.int64)

ARGS = {
    'y_true': y_true,
    'y_pred': y_pred
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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
cluster_acc Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
BGU-CS-VIL/DeepDPM/src/utils.py
code served (permissive licence) · get_code("b87c44077cdbce2d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
from scipy.optimize import linear_sum_assignment as linear_assignment

np.random.seed(0)

# Create small sample data for clustering
# y_true: ground truth cluster labels
# y_pred: predicted cluster labels

y_true = np.array([0, 0, 1, 1, 2, 2], dtype=np.int64)
y_pred = np.array([0, 0, 2, 2, 1, 1], dtype=np.int64)

ARGS = {
    "y_true": y_true,
    "y_pred": y_pred,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
collate_custom Ran
ran, violated the contract we drafted
BGU-CS-VIL/DeepDPM/src/get_embbedings/imagenet.py
code served (permissive licence) · get_code("3fc4a5ed41115671")
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 3, dtype family float, shape (3, 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
import collections

torch.manual_seed(0)

# Test with different batch types
# Case 1: batch of torch.Tensor
batch_tensors = [torch.randn(2, 4) for _ in range(3)]

# Case 2: batch of numpy arrays
batch_numpy = [np.random.randn(2, 4) for _ in range(3)]

# Case 3: batch of integers
batch_ints = [1, 2, 3, 4]

# Case 4: batch of floats
batch_floats = [1.5, 2.5, 3.5]

# Case 5: batch of strings
batch_strings = ["hello", "world", "test"]

# Case 6: batch of dicts (Mapping)
batch_dicts = [
    {"key1": torch.randn(2, 4), "key2": 5, "idx_field": 999},
    {"key1": torch.randn(2, 4), "key2": 6, "idx_field": 1000},
]

# Case 7: batch of np.int64
batch_np_int64 = [np.int64(10), np.int64(20), np.int64(30)]

# Case 8: batch of sequences (lists/tuples)
batch_sequences = [
    (torch.randn(2, 4), 5),
    (torch.randn(2, 4), 6),
]

# Choose one batch for the ARGS
ARGS = {
    "batch": batch_tensors
}
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 3, dtype family float, shape (3, 2, 4) — shape (3, 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_mock_dataset Ran
ran; our contract draft was wrong, not the code
BGU-CS-VIL/DeepDPM/src/datasets.py
code served (permissive licence) · get_code("38f82d3587625eba")
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 TensorDataset. 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 TensorDataset

torch.manual_seed(0)

ARGS = {
    "dim": (4, 3),
    "len": 2,
    "dtype": torch.float32
}
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 TensorDataset — a TensorDataset 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
generator Ran
ran; our contract draft was wrong, not the code · constructed an object; no result was computed
XifengGuo/ASPC-DA/ASPC.py
pointer only (licence: NONE) · get_code("783463d4dbb4faa6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, XifengGuo/ASPC-DA 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.

load_pendigits Ran
ran; our contract draft was wrong, not the code
XifengGuo/DEC-keras/datasets.py
code served (permissive licence) · get_code("bb97a52065e28cef")
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 os
import tempfile
import numpy as np

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

# Create mock pendigits data files
os.makedirs(data_path, exist_ok=True)

# Create pendigits.tra (training data)
# Format: 16 features + 1 label per line
train_data = []
for i in range(100):
    features = [str(np.random.randint(0, 100)) for _ in range(16)]
    label = str(np.random.randint(0, 10))
    train_data.append(','.join(features + [label]))

with open(os.path.join(data_path, 'pendigits.tra'), 'w') as f:
    f.write('\n'.join(train_data))

# Create pendigits.tes (test data)
test_data = []
for i in range(50):
    features = [str(np.random.randint(0, 100)) for _ in range(16)]
    label = str(np.random.randint(0, 10))
    test_data.append(','.join(features + [label]))

with open(os.path.join(data_path, 'pendigits.tes'), 'w') as f:
    f.write('\n'.join(test_data))

# Create pendigits.names (metadata file - not strictly used by function)
with open(os.path.join(data_path, 'pendigits.names'), 'w') as f:
    f.write('Pendigits dataset\n')

ARGS = {
    'data_path': data_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 — 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
load_usps Ran
ran; our contract draft was wrong, not the code
XifengGuo/DEC-keras/datasets.py
code served (permissive licence) · get_code("f92760073da5ad67")
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 os
import tempfile
import numpy as np

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

# Create mock USPS data files
train_data = """jf
1 0.1 0.2 0.3 0.4
2 0.5 0.6 0.7 0.8
3 0.9 1.0 1.1 1.2
9 0.2 0.3 0.4 0.5
"""

test_data = """jf
4 0.1 0.2 0.3 0.4
5 0.5 0.6 0.7 0.8
9 0.9 1.0 1.1 1.2
"""

train_file = os.path.join(temp_dir, 'usps_train.jf')
test_file = os.path.join(temp_dir, 'usps_test.jf')

with open(train_file, 'w') as f:
    f.write(train_data)

with open(test_file, 'w') as f:
    f.write(test_data)

ARGS = {
    'data_path': temp_dir
}
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 — 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
merge_datasets Ran
ran; our contract draft was wrong, not the code
BGU-CS-VIL/DeepDPM/src/datasets.py
code served (permissive licence) · get_code("72b10f5c47341c58")
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 ConcatDataset. 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 TensorDataset

torch.manual_seed(0)

# Create two small TensorDatasets
# TensorDataset takes tensors and zips them together
# We'll create datasets with 2 samples each, with feature dimension 4

# First dataset: 2 samples, 4 features per sample
x1 = torch.randn(2, 4)
y1 = torch.randint(0, 2, (2,))
set_1 = TensorDataset(x1, y1)

# Second dataset: 2 samples, 4 features per sample
x2 = torch.randn(2, 4)
y2 = torch.randint(0, 2, (2,))
set_2 = TensorDataset(x2, y2)

ARGS = {
    "set_1": set_1,
    "set_2": set_2,
}
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 ConcatDataset — a ConcatDataset 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
run_on_embeddings_hyperparams Ran
ran; our contract draft was wrong, not the code
BGU-CS-VIL/DeepDPM/DeepDPM.py
code served (permissive licence) · get_code("2090189ae1956d7c")
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 ArgumentParser. 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 argparse import ArgumentParser

torch.manual_seed(0)

# Create a minimal parent parser
parent_parser = ArgumentParser(add_help=False)
parent_parser.add_argument('--some_parent_arg', type=str, default='parent_value')

# Define ARGS dict with all parameters for run_on_embeddings_hyperparams
ARGS = {
    'parent_parser': parent_parser
}
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 ArgumentParser — a ArgumentParser output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ArgumentParser 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
transform_embeddings Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
BGU-CS-VIL/DeepDPM/src/datasets.py
code served (permissive licence) · get_code("b13af437306e5c2a")
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
from sklearn.preprocessing import MinMaxScaler, StandardScaler, Normalizer

torch.manual_seed(0)

# Create small sample embedding data (2 samples, 4 features)
# numpy array is what sklearn transformers expect
data = np.random.randn(2, 4).astype(np.float32)

ARGS = {
    'transform_type': 'normalize',
    'data': data
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn, 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
ASPC Not yet run
unverified (harvested, no recorded run)
XifengGuo/ASPC-DA/ASPC.py
pointer only (licence: NONE) · get_code("5f12b94d46b956f7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, XifengGuo/ASPC-DA 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. No invocation recorded.

EDESC Not yet run
unverified (harvested, no recorded run)
JinyuCai95/EDESC-pytorch/EDESC.py
code served (permissive licence) · get_code("051c59fb408cf6dc")
How it ran

no invocation recorded.

MyImageGenerator Not yet run
unverified (harvested, no recorded run)
XifengGuo/ASPC-DA/ASPC.py
pointer only (licence: NONE) · get_code("ce4d615ade3cc0af")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, XifengGuo/ASPC-DA 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. No invocation recorded.

acc Not yet run
unverified (harvested, no recorded run)
XifengGuo/ASPC-DA/ASPC.py
pointer only (licence: NONE) · get_code("1386e9795c359688")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, XifengGuo/ASPC-DA 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. No invocation recorded.

extract_vgg16_features Not yet run
unverified (harvested, no recorded run)
XifengGuo/DEC-keras/datasets.py
code served (permissive licence) · get_code("86b45a82862b929d")
How it ran

no invocation recorded.

parse_minimal_args Not yet run
unverified (harvested, no recorded run)
BGU-CS-VIL/DeepDPM/DeepDPM.py
code served (permissive licence) · get_code("0bb5af34445d9f1a")
How it ran

no invocation recorded.

pretrain_ae Not yet run
unverified (harvested, no recorded run)
JinyuCai95/EDESC-pytorch/EDESC.py
code served (permissive licence) · get_code("7b2bc9b695748564")
How it ran

no invocation recorded.

Also present in other papers' code

Of 23 samples lifted from this paper's linked repositories, 23 carry a function identity and 0 do not. Of the 23 distinct functions with an identity, 5 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 0 later) and 18 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
acc 2 other papers
2 dated earlier
1511.06335 (2015-11)
Unsupervised Deep Embedding for Clustering Analysis
1907.00481 (2019-07)
Spectral Clustering with Graph Neural Networks for Graph Pooling
autoencoder 1 other paper
1 dated earlier
2401.15989 (2024-01)
Deep Embedding Clustering Driven by Sample Stability
generate_mock_dataset 1 other paper
1 dated earlier
2408.13492 (2024-08)
Online Continuous Generalized Category Discovery
merge_datasets 1 other paper
1 dated earlier
2408.13492 (2024-08)
Online Continuous Generalized Category Discovery
transform_embeddings 1 other paper
1 dated earlier
2408.13492 (2024-08)
Online Continuous Generalized Category Discovery

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

Image clustering is a crucial but challenging task in multimedia machine learning. Recently the combination of clustering with deep learning has achieved promising performance against conventional methods on high-dimensional image data. Unfortunately, existing deep clustering methods (DC) often ignore the importance of information fusion with a global perception field among different image regions on clustering images, especially complex ones. Additionally, the learned features are usually clustering-unfriendly in terms of dimensionality and are based only on simple distance information for the clustering. In this regard, we propose a deep embedded image clustering TDEC, which for the first time to our knowledge, jointly considers feature representation, dimensional preference, and robust assignment for image clustering. Specifically, we introduce the Transformer to form a novel module T-Encoder to learn discriminative features with global dependency while using the Dim-Reduction block to build a friendly low-dimensional space favoring clustering. Moreover, the distribution information of embedded features is considered in the clustering process to provide reliable supervised signals for joint training. Our method is robust and allows for more flexibility in data size, the number of clusters, and the context complexity. More importantly, the clustering performance of TDEC is much higher than recent competitors. Extensive experiments with state-of-the-art approaches on complex datasets show the superiority of TDEC. • Computing methodologies → Computer vision.

For agents

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

get_harvested_code_for_paper("2603.26746")
get_code_for_paper("2603.26746")
have("2603.26746")

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

curl https://syntology.ai/api/ran/2603.26746.json

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

[![Syntology run record](https://syntology.ai/api/ran/2603.26746.svg)](https://syntology.ai/paper/2603.26746)

Connect an agent — have() is free.