SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.10396 · 2020

NSGANetV2: Evolutionary Multi-Objective Surrogate-Assisted Neural Architecture Search

arXiv · PDF · Open in the Atlas

Code that ran

4 of 8 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, 1 violated it and 2 have no contract verdict; 1 failure of our instrument; 4 no recorded run), measured 2026-11-21; 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/2007.10396.json · badge: /api/ran/2007.10396.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
mikelzc1990/nsganetv2 canonical (classifier label)
classifier basis: census
4 ran · 4 no recorded run
FunctionStatusWhere it lives
get_correlation Ran
ran on a synthesized input
mikelzc1990/nsganetv2/utils.py
code served (permissive licence) · get_code("285dcbbdea2ec2b3")
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
import torch

torch.manual_seed(0)

# Create small numpy arrays for prediction and target
# The function performs element-wise operations and statistical correlations
# so we need arrays with some variability

np.random.seed(0)
prediction = np.random.randn(8)
target = np.random.randn(8)

ARGS = {
    "prediction": prediction,
    "target": target,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy, 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
look_up_latency Ran
ran on a synthesized input
mikelzc1990/nsganetv2/utils.py
code served (permissive licence) · get_code("90e3cacca1501e59")
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
import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a minimal LUT (Lookup Table) class that implements the predict method
class MockLUT:
    def predict(self, layer_name, input_shape, output_shape, expand=None, kernel=None, stride=None, idskip=None, se=None):
        # Return a mock latency value
        return float(np.prod(output_shape) if output_shape else 1.0)

# Create a minimal block config
def create_block_config(in_channels, out_channels, expand_ratio=6, kernel_size=3, stride=1, use_se=False, shortcut=True):
    return {
        'shortcut': 'add' if shortcut else None,
        'mobile_inverted_conv': {
            'in_channels': in_channels,
            'out_channels': out_channels,
            'expand_ratio': expand_ratio,
            'kernel_size': kernel_size,
            'stride': stride,
            'use_se': use_se
        }
    }

# Create a minimal MBConv block
class MBConvBlock(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.config = config

# Create a minimal network matching the function's expectations
class MockEfficientNet(nn.Module):
    def __init__(self):
        super().__init__()
        
        # first_conv layer
        self.first_conv = nn.Conv2d(3, 32, kernel_size=3, stride=2, padding=1)
        self.first_conv.out_channels = 32
        
        # Create some blocks with different strides
        self.blocks = nn.ModuleList([
            MBConvBlock(create_block_config(32, 16, expand_ratio=1, stride=1, use_se=False, shortcut=True)),
            MBConvBlock(create_block_config(16, 24, expand_ratio=6, stride=2, use_se=True, shortcut=True)),
            MBConvBlock(create_block_config(24, 40, expand_ratio=6, stride=2, use_se=True, shortcut=True)),
            MBConvBlock(create_block_config(40, 80, expand_ratio=6, stride=2, use_se=True, shortcut=False)),
        ])
        
        # final_expand_layer
        self.final_expand_layer = nn.Linear(80, 320)
        self.final_expand_layer.in_channels = 80
        self.final_expand_layer.out_channels = 320
        
        # feature_mix_layer
        self.feature_mix_layer = nn.Linear(320, 1280)
        self.feature_mix_layer.in_channels = 320
        self.feature_mix_layer.out_channels = 1280
        
        # classifier
        self.classifier = nn.Linear(1280, 1000)
        self.classifier.in_features = 1280
        self.classifier.out_features = 1000

# Instantiate the network and LUT
net = MockEfficientNet()
lut = MockLUT()

# Define ARGS
ARGS = {
    'net': net,
    'lut': lut,
    'resolution': 224
}
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 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
pad_none Ran
ran; our contract draft was wrong, not the code
mikelzc1990/nsganetv2/evaluator.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from CownowAn/DiffusionNAG, is pointer only (licence: NONE) · get_code("6db40799d999d837")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from CownowAn/DiffusionNAG, 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.

parse_string_list Ran
ran, violated the contract we drafted
mikelzc1990/nsganetv2/evaluator.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from CownowAn/DiffusionNAG, is pointer only (licence: NONE) · get_code("59ef743ae49b58e1")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from CownowAn/DiffusionNAG, 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.

get_acc_predictor Not yet run
unverified (harvested, no recorded run)
mikelzc1990/nsganetv2/acc_predictor/factory.py
code served (permissive licence) · get_code("ede09871c3c8b79b")
How it ran

no invocation recorded.

infer Not yet run
unverified (harvested, no recorded run)
mikelzc1990/nsganetv2/train_cifar.py
code served (permissive licence) · get_code("2d6dc16493521afd")
How it ran

no invocation recorded.

train Not yet run
unverified (harvested, no recorded run)
mikelzc1990/nsganetv2/train_cifar.py
code served (permissive licence) · get_code("22ac9a49daa933c8")
How it ran

no invocation recorded.

validate Not yet run
unverified (harvested, no recorded run)
mikelzc1990/nsganetv2/train_imagenet.py
code served (permissive licence) · get_code("b5886490f0a3569d")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 3 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 2 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
infer 1 other paper
1 dated earlier
2003.13880 (2020-03)
MUXConv: Information Multiplexing in Convolutional Neural Networks
pad_none 1 other paper
1 later
2305.16943 (2023-05)
DiffusionNAG: Predictor-guided Neural Architecture Generation with Diffusion Models
parse_string_list 1 other paper
1 later
2305.16943 (2023-05)
DiffusionNAG: Predictor-guided Neural Architecture Generation with Diffusion Models

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 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

In this paper, we propose an efficient NAS algorithm for generating task-specific models that are competitive under multiple competing objectives. It comprises of two surrogates, one at the architecture level to improve sample efficiency and one at the weights level, through a supernet, to improve gradient descent training efficiency. On standard benchmark datasets (C10, C100, ImageNet), the resulting models, dubbed NSGANetV2, either match or outperform models from existing approaches with the search being orders of magnitude more sample efficient. Furthermore, we demonstrate the effectiveness and versatility of the proposed method on six diverse non-standard datasets, e.g. STL-10, Flowers102, Oxford Pets, FGVC Aircrafts etc. In all cases, NSGANetV2s improve the state-of-the-art (under mobile setting), suggesting that NAS can be a viable alternative to conventional transfer learning approaches in handling diverse scenarios such as small-scale or fine-grained datasets. Code is available at https://github.com/mikelzc1990/nsganetv2

For agents

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

get_harvested_code_for_paper("2007.10396")
get_code_for_paper("2007.10396")
have("2007.10396")

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

curl https://syntology.ai/api/ran/2007.10396.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.10396.svg)](https://syntology.ai/paper/2007.10396)

Connect an agent — have() is free.