SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2108.10310 · ICCV · 2021

Ranking Models in Unlabeled New Environments

Hongdong Li, Liang Zheng, Weijian Deng, Xiaoxiao Sun, Yunzhong Hou

arXiv · PDF · Open in the Atlas

Code that ran

6 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 6 have no contract verdict; 0 failures of our instrument; 3 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/2108.10310.json · badge: /api/ran/2108.10310.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
sxzrt/Proxy-Set canonical (classifier label)
classifier basis: census
5 ran · 3 no recorded run
sxzrt/semi-structured-dataset-representations community
Papers with Code archive
1 ran · 0 no recorded run
FunctionStatusWhere it lives
Set_Rpresentation Ran
ran on a synthesized input · constructed an object; no result was computed
sxzrt/semi-structured-dataset-representations/Set_rep/regression_new.py
code served (permissive licence) · get_code("b05adfaeefca09cd")
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 an instance of its own class, Set_Rpresentation: the run constructed the object and did not call it, so no result was computed. 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
import torch.utils.data as data
import tempfile
import os

torch.manual_seed(0)
np.random.seed(0)

# Create temporary directory and mock numpy files
tmpdir = tempfile.mkdtemp()

# Create mock .npy files for indices 0 and 1
for idx in [0, 1]:
    np.save(os.path.join(tmpdir, f'mock_{idx}_shape.npy'), np.random.randn(1, 30, 64).astype(np.float32))
    np.save(os.path.join(tmpdir, f'mock_{idx}_cluster.npy'), np.random.randn(1, 10, 64).astype(np.float32))
    np.save(os.path.join(tmpdir, f'mock_{idx}_fps.npy'), np.random.randn(1, 100, 64).astype(np.float32))
    np.save(os.path.join(tmpdir, f'mock_{idx}_mean.npy'), np.random.randn(64).astype(np.float32))

path = os.path.join(tmpdir, 'mock')
data_indices = [0, 1]
labels = np.array([0.0, 1.0])

ARGS = {
    'path': path,
    'data': data_indices,
    'label': labels,
    'transform': None,
    'target_transform': None
}
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 Set_Rpresentation — the run constructed an instance of its own class and did not call it; no shape 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
calculate_frechet_distance Ran
ran on a synthesized input
sxzrt/Proxy-Set/domain_gap/fd_score_label.py
code served (permissive licence) · get_code("e7839bb8b15f8f40")
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

np.random.seed(0)

# Create sample data for the function
# The function expects lists of means and covariances for two distributions

# Dimension of feature space
feature_dim = 4

# Number of layers/comparisons
num_layers = 2

# Create means (feature vectors)
mu1_list = [np.random.randn(feature_dim) for _ in range(num_layers)]
mu2_list = [np.random.randn(feature_dim) for _ in range(num_layers)]

# Create positive semi-definite covariance matrices
sigma1_list = []
sigma2_list = []

for _ in range(num_layers):
    # Generate random matrix and make it positive semi-definite via A^T @ A
    A1 = np.random.randn(feature_dim, feature_dim)
    sigma1_list.append(A1 @ A1.T)
    
    A2 = np.random.randn(feature_dim, feature_dim)
    sigma2_list.append(A2 @ A2.T)

ARGS = {
    'mu1_list': mu1_list,
    'sigma1_list': sigma1_list,
    'mu2_list': mu2_list,
    'sigma2_list': sigma2_list,
    'eps': 1e-06
}
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
make_square Ran
ran on a synthesized input
sxzrt/Proxy-Set/domain_gap/fd_score.py
code served (permissive licence) · get_code("898416037b5bc82f")
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 3, dtype family int, shape (4, 4, 2) 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

# Create a small test image with shape (height, width, channels)
# Using distinct dimensions: 4 height, 3 width, 2 channels
test_image = np.random.randint(0, 256, size=(4, 3, 2), dtype=np.uint8)

ARGS = {
    "image": test_image,
    "max_dim": 512
}
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 3, dtype family int, shape (4, 4, 2) — shape (4, 4, 2) 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
polynomial_mmd Ran
ran on a synthesized input
sxzrt/Proxy-Set/domain_gap/kd_score.py
code served (permissive licence) · get_code("7405f2156522f874")
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
from sklearn.metrics.pairwise import polynomial_kernel

np.random.seed(0)

# Create small sample data
codes_g = np.random.randn(4, 5).astype(np.float32)
codes_r = np.random.randn(4, 5).astype(np.float32)

ARGS = {
    'codes_g': codes_g,
    'codes_r': codes_r,
    'degree': 3,
    'gamma': None,
    'coef0': 1,
    'var_at_m': None,
    'ret_var': True,
}
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 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
random_apply Ran
ran on a synthesized input
sxzrt/Proxy-Set/domain_gap/DG_gegerate_data.py
code served (permissive licence) · get_code("d04f187058cb2a9c")
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 SymbolicTensor. 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 tensorflow.compat.v1 as tf

tf.disable_v2_behavior()

def simple_transform(x):
    """Simple transformation function that doubles the input."""
    return x * 2

# Create a small tensor
x = tf.constant([1.0, 2.0, 3.0, 4.0], dtype=tf.float32)

# Define probability
p = 0.7

# Define function to apply
func = simple_transform

ARGS = {
    "func": func,
    "p": p,
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SymbolicTensor — a SymbolicTensor 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
random_brightness Ran
ran on a synthesized input
sxzrt/Proxy-Set/domain_gap/DG_gegerate_data.py
code served (permissive licence) · get_code("e1aaf77a8784526a")
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 SymbolicTensor. 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 tensorflow.compat.v1 as tf
import numpy as np

tf.disable_v2_behavior()

# Create a small image tensor
# Images are typically [height, width, channels]
image = tf.constant(np.random.rand(4, 3, 3).astype(np.float32))

# max_delta is a float controlling the brightness change range
max_delta = 0.2

# impl is a string specifying the implementation
impl = 'simclrv2'

ARGS = {
    'image': image,
    'max_delta': max_delta,
    'impl': impl
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SymbolicTensor — a SymbolicTensor 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
calculate_frechet_distance Not yet run
unverified (harvested, no recorded run)
sxzrt/Proxy-Set/domain_gap/fd_score.py
code served (permissive licence) · get_code("0def50a351111624")
How it ran

no invocation recorded. this sample carries JSON test cases instead of a setup block; get_reference_implementation serves them with the level they back.

get_activations Not yet run
unverified (harvested, no recorded run)
sxzrt/Proxy-Set/domain_gap/fd_score_label.py
code served (permissive licence) · get_code("8b0372a487a32700")
How it ran

no invocation recorded.

to_grayscale Not yet run
unverified (harvested, no recorded run)
sxzrt/Proxy-Set/domain_gap/DG_gegerate_data.py
code served (permissive licence) · get_code("49c827442d2fb761")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 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, 5 later) and 4 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
calculate_frechet_distance 74 other papers
20 dated earlier, 53 later, 1 same year, order not determinable
1610.09585 (2016-10)
Conditional Image Synthesis With Auxiliary Classifier GANs
1706.08500 (2017-06)
GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
1807.11458 (2018-07)
To learn image super-resolution, use a GAN to learn how to do image degradation first
and 71 more not named here
random_apply 4 other papers
2 dated earlier, 2 later
2006.05576 (2020-06)
Self-supervised Learning from a Multi-view Perspective
2105.12723 (2021-05)
Nested Hierarchical Transformer: Towards Accurate, Data-Efficient and Interpretable Visual Understanding
2112.11435 (2021-12)
Learned Queries for Efficient Local Attention
and 1 more not named here
to_grayscale 4 other papers
2 dated earlier, 2 later
2006.05576 (2020-06)
Self-supervised Learning from a Multi-view Perspective
2105.12723 (2021-05)
Nested Hierarchical Transformer: Towards Accurate, Data-Efficient and Interpretable Visual Understanding
2109.12909 (2021-09)
Compressive Visual Representations
and 1 more not named here
polynomial_mmd 3 other papers
1 dated earlier, 2 later
2003.05033 (2020-03)
Generalized Energy Based Models
2210.16559 (2022-10)
Few-shot Image Generation via Adaptation-Aware Kernel Modulation
2304.07574 (2023-04)
Exploring Incompatible Knowledge Transfer in Few-shot Image Generation
random_brightness 2 other papers
1 dated earlier, 1 later
2105.12723 (2021-05)
Nested Hierarchical Transformer: Towards Accurate, Data-Efficient and Interpretable Visual Understanding
2112.11435 (2021-12)
Learned Queries for Efficient Local Attention

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

Consider a scenario where we are supplied with a number of ready-to-use models trained on a certain source domain and hope to directly apply the most appropriate ones to different target domains based on the models' relative performance. Ideally we should annotate a validation set for model performance assessment on each new target environment, but such annotations are often very expensive. Under this circumstance, we introduce the problem of ranking models in unlabeled new environments. For this problem, we propose to adopt a proxy dataset that 1) is fully labeled and 2) well reflects the true model rankings in a given target environment, and use the performance rankings on the proxy sets as surrogates. We first select labeled datasets as the proxy. Specifically, datasets that are more similar to the unlabeled target domain are found to better preserve the relative performance rankings. Motivated by this, we further propose to search the proxy set by sampling images from various datasets that have similar distributions as the target. We analyze the problem and its solutions on the person re-identification (re-ID) task, for which sufficient datasets are publicly available, and show that a carefully constructed proxy set effectively captures relative performance ranking in new environments. Code is avalible at https://github.com/sxzrt/Proxy-Set.

For agents

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

get_harvested_code_for_paper("2108.10310")
get_code_for_paper("2108.10310")
have("2108.10310")

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

curl https://syntology.ai/api/ran/2108.10310.json

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

[![Syntology run record](https://syntology.ai/api/ran/2108.10310.svg)](https://syntology.ai/paper/2108.10310)

Connect an agent — have() is free.