SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2202.06866 · ICLR · 2022

Delaunay Component Analysis for Evaluation of Data Representations

Petra Poklukar, Danica Kragic, Vladislav Polianskii, Anastasia Varava, Florian Pokorny

arXiv · PDF · Open in the Atlas

Code that ran

7 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 7 have no contract verdict; 0 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2202.06866.json · badge: /api/ran/2202.06866.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
petrapoklukar/dca canonical (classifier label)
classifier basis not recorded on the link
7 ran · 0 no recorded run
FunctionStatusWhere it lives
get_color Ran
ran on a synthesized input
petrapoklukar/dca/dca/visualization.py
code served (permissive licence) · get_code("30b6fc26960f032b")
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 torch

torch.manual_seed(0)

# Construct valid arguments for get_color
edge = [0, 5]  # A list of two indices
num_R = 3      # Number of R points in the graph

ARGS = {
    'edge': edge,
    'num_R': num_R
}
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 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
get_parameters Ran
ran on a synthesized input
petrapoklukar/dca/dca/loggers.py
code served (permissive licence) · get_code("84664625dedeb394")
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 torch.nn as nn

# Create a simple test class with various attributes
class TestClass:
    class_int = 42
    class_str = "test_string"
    class_float = 3.14
    class_list = [1, 2, 3]
    class_bool = True
    class_tensor = torch.tensor([1.0, 2.0])  # This won't be included (not primitive type)
    
    def class_method(self):
        pass

ARGS = {
    "class_name": TestClass
}
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 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
get_pruned_representation_idxs_by_class Ran
ran on a synthesized input
petrapoklukar/dca/experiments/contrastive_learning/CL_utils.py
code served (permissive licence) · get_code("5ab465dfd8afcea8")
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 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 numpy as np

# Create a labels array with different class values
# We'll have 100 samples with 3 different classes (0, 1, 2)
labels = np.array([0] * 40 + [1] * 35 + [2] * 25)

# Shuffle to make it more realistic
np.random.seed(42)
np.random.shuffle(labels)

# Define the R classes to extract
Rclasses = [0, 1, 2]

# Define which classes to prune
classes_to_prune = [1]

# Percentage of points to discard (e.g., 30%)
perc_to_discard = 0.3

ARGS = {
    "labels": labels,
    "Rclasses": Rclasses,
    "classes_to_prune": classes_to_prune,
    "perc_to_discard": perc_to_discard,
}
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 int — 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
get_representation_idxs_by_class Ran
ran on a synthesized input
petrapoklukar/dca/experiments/contrastive_learning/CL_utils.py
code served (permissive licence) · get_code("b8a9dd388787d96a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 1, dtype family int, shape (7,) 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 up arguments for get_representation_idxs_by_class
labels = np.array([0, 1, 0, 2, 1, 2, 0, 1, 2, 0])
Rclasses = [0, 2]

ARGS = {
    "labels": labels,
    "Rclasses": Rclasses,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family int, shape (7,) — shape (7,) 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_representations_by_class Ran
ran on a synthesized input
petrapoklukar/dca/experiments/contrastive_learning/CL_utils.py
code served (permissive licence) · get_code("d5900b018d972545")
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

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

# Create small sample representations (batch_size=10, feature_dim=4)
representations = np.random.randn(10, 4).astype(np.float32)

# Create corresponding labels (5 samples of class 0, 3 of class 1, 2 of class 2)
labels = np.array([0, 0, 0, 0, 0, 1, 1, 1, 2, 2], dtype=np.int32)

# Define R classes to extract
Rclasses = [0, 1]

# Define E classes to extract (optional)
Eclasses = [2]

# Construct the arguments dictionary
ARGS = {
    "representations": representations,
    "labels": labels,
    "Rclasses": Rclasses,
    "Eclasses": Eclasses,
}
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
logger_time Ran
ran on a synthesized input
petrapoklukar/dca/dca/delaunay_graph.py
code served (permissive licence) · get_code("24f72f380835031d")
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 function. 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 time
import logging
from functools import wraps

logger = logging.getLogger('DCA_info_logger')
result_logger = logging.getLogger('DCA_result_logger')

def logger_time(func):
    """Prints the execution time for the decorated function."""

    @wraps(func)
    def wrapper(*args, **kwargs):
        start = time.time()
        result = func(*args, **kwargs)
        end = time.time()
        logger.debug('Function {0} executed in {1:f} s'.format(func.__name__, end - start))
        result_logger.debug('Elapsed time: {0}-{1:f} s'.format(func.__name__, end - start))
        return result
    return wrapper


def sample_function(x, y):
    """A simple function to decorate with logger_time."""
    return x + y


decorated_function = logger_time(sample_function)

ARGS = {
    'func': decorated_function
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
logger_time Ran
ran on a synthesized input
petrapoklukar/dca/dca/loggers.py
code served (permissive licence) · get_code("7bac25afd5656c90")
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 function. 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 logging
import time
from functools import wraps

# Set up logging
logging.basicConfig(level=logging.DEBUG)
logger = logging.getLogger('DCA_info_logger')
time_logger = logging.getLogger('DCA_time_logger')

# Define the decorator
def logger_time(func):
    """Prints the execution time for the decorated function."""

    @wraps(func)
    def wrapper(*args, **kwargs):
        start = time.time()
        result = func(*args, **kwargs)
        end = time.time()
        logger.debug('Function {0} executed in {1:f} s'.format(func.__name__, end - start))
        time_logger.debug('Elapsed time: {0}-{1:f} s'.format(func.__name__, end - start))
        return result
    return wrapper

# Create a simple test function to decorate
@logger_time
def sample_function(x, y=10):
    """A simple function to test the decorator."""
    time.sleep(0.01)
    return x + y

# Build ARGS dict with all parameters the decorator needs
ARGS = {
    'func': sample_function
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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 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

Advanced representation learning techniques require reliable and general evaluation methods. Recently, several algorithms based on the common idea of geometric and topological analysis of a manifold approximated from the learned data representations have been proposed. In this work, we introduce Delaunay Component Analysis (DCA) -an evaluation algorithm which approximates the data manifold using a more suitable neighbourhood graph called Delaunay graph. This provides a reliable manifold estimation even for challenging geometric arrangements of representations such as clusters with varying shape and density as well as outliers, which is where existing methods often fail. Furthermore, we exploit the nature of Delaunay graphs and introduce a framework for assessing the quality of individual novel data representations. We experimentally validate the proposed DCA method on representations obtained from neural networks trained with contrastive objective, supervised and generative models, and demonstrate various use cases of our extended single point evaluation framework. Recently, more general evaluation methods such as Geometry Score (GS) (Khrulkov & Oseledets, 2018), Improved Precision and Recall (IPR) (Kynkäänniemi et al., 2019) and Geometric Component Analysis (GeomCA) (Poklukar et al., 2021) have been proposed.

For agents

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

get_harvested_code_for_paper("2202.06866")
get_code_for_paper("2202.06866")
have("2202.06866")

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

curl https://syntology.ai/api/ran/2202.06866.json

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

[![Syntology run record](https://syntology.ai/api/ran/2202.06866.svg)](https://syntology.ai/paper/2202.06866)

Connect an agent — have() is free.