Layer-specific feature preference in a trained AlexNet model revealed by stylized image analysis.
3 blocks of code here have a permanent identifier (oscr:b:…), listed with the matches below: citable for good, with a page of its own. Select a match and press Cite to copy it. What a block identifier is.
The 3 matches
- [1] § Methods › AlexNet of the DCNN model for classifying objects ↔ gcn/main.py, lines 67 to 112 · score 0.67 · dropout layers, activation function, ReLU, architecture, probability, classes · block
oscr:b:AQBRAP51PY37V-N· Report a mismatch - [2] § Methods › AlexNet of the DCNN model for classifying objects ↔ torch_to_pytorch.py, lines 153 to 240 · score 0.52 · max pooling, ReLU, PyTorch, dropout, Conv · block
oscr:b:B0D9P7D5HDGT6-X· Report a mismatch - [3] § Methods › Training the AlexNet model using the ImageNet dataset ↔ imagenet/main.py, lines 199 to 283 · score 0.51 · cross entropy loss, validate, optimization, batch, ImageNet, trained · block
oscr:b:CRN9P4V06AT4W-T· Report a mismatch
Paper
Loaded from Europe PMC by your browser; OSCR does not store it: doi.org · Europe PMC
The paper is loaded when this pane is shown.
The authors' code
Python · 256 lines · 11 KB · BSD-3-Clause · 1 match
- import os
- import time
- import requests
- import tarfile
- import numpy as np
- import argparse
- import torch
- from torch import nn
- import torch.nn.functional as F
- from torch.optim import Adam
- class GraphConv(nn.Module):
- """
- Graph Convolutional Layer described in "Semi-Supervised Classification with Graph Convolutional Networks".
- Given an input feature representation for each node in a graph, the Graph Convolutional Layer aims to aggregate
- information from the node's neighborhood to update its own representation. This is achieved by applying a graph
- convolutional operation that combines the features of a node with the features of its neighboring nodes.
- Mathematically, the Graph Convolutional Layer can be described as follows:
- H' = f(D^(-1/2) * A * D^(-1/2) * H * W)
- where:
- H: Input feature matrix with shape (N, F_in), where N is the number of nodes and F_in is the number of
- input features per node.
- A: Adjacency matrix of the graph with shape (N, N), representing the relationships between nodes.
- W: Learnable weight matrix with shape (F_in, F_out), where F_out is the number of output features per node.
- D: The degree matrix.
- """
- def __init__(self, input_dim, output_dim, use_bias=False):
- super(GraphConv, self).__init__()
- # Initialize the weight matrix W (in this case called `kernel`)
- self.kernel = nn.Parameter(torch.Tensor(input_dim, output_dim))
- nn.init.xavier_normal_(self.kernel) # Initialize the weights using Xavier initialization
- # Initialize the bias (if use_bias is True)
- self.bias = None
- if use_bias:
- self.bias = nn.Parameter(torch.Tensor(output_dim))
- nn.init.zeros_(self.bias) # Initialize the bias to zeros
- def forward(self, input_tensor, adj_mat):
- """
- Performs a graph convolution operation.
- Args:
- input_tensor (torch.Tensor): Input tensor representing node features.
- adj_mat (torch.Tensor): Normalized adjacency matrix representing graph structure.
- Returns:
- torch.Tensor: Output tensor after the graph convolution operation.
- """
- support = torch.mm(input_tensor, self.kernel) # Matrix multiplication between input and weight matrix
- output = torch.spmm(adj_mat, support) # Sparse matrix multiplication between adjacency matrix and support
- # Add the bias (if bias is not None)
- if self.bias is not None:
- output = output + self.bias
- return output
- class GCN(nn.Module):
- """
- Graph Convolutional Network (GCN) as described in the paper `"Semi-Supervised Classification with Graph
- Convolutional Networks" <https://arxiv.org/pdf/1609.02907.pdf>`.
- The Graph Convolutional Network is a deep learning architecture designed for semi-supervised node
- classification tasks on graph-structured data. It leverages the graph structure to learn node representations
- by propagating information through the graph using graph convolutional layers.
- The original implementation consists of two stacked graph convolutional layers. The ReLU activation function is
- applied to the hidden representations, and the Softmax activation function is applied to the output representations.
- """
- def __init__(self, input_dim, hidden_dim, output_dim, use_bias=True, dropout_p=0.1):
- super(GCN, self).__init__()
- # Define the Graph Convolution layers
- self.gc1 = GraphConv(input_dim, hidden_dim, use_bias=use_bias)
- self.gc2 = GraphConv(hidden_dim, output_dim, use_bias=use_bias)
- # Define the dropout layer
- self.dropout = nn.Dropout(dropout_p)
- def forward(self, input_tensor, adj_mat):
- """
- Performs forward pass of the Graph Convolutional Network (GCN).
- Args:
- input_tensor (torch.Tensor): Input node feature matrix with shape (N, input_dim), where N is the number of nodes
- and input_dim is the number of input features per node.
- adj_mat (torch.Tensor): Normalized adjacency matrix of the graph with shape (N, N), representing the relationships between
- nodes.
- Returns:
- torch.Tensor: Output tensor with shape (N, output_dim), representing the predicted class probabilities for each node.
- """
- # Perform the first graph convolutional layer
- x = self.gc1(input_tensor, adj_mat)
- x = F.relu(x) # Apply ReLU activation function
- x = self.dropout(x) # Apply dropout regularization
- # Perform the second graph convolutional layer
- x = self.gc2(x, adj_mat)
- # Apply log-softmax activation function for classification
- return F.log_softmax(x, dim=1)
- def load_cora(path='./cora', device='cpu'):
- """
- The graph convolutional operation rquires the normalized adjacency matrix: D^(-1/2) * A * D^(-1/2). This step
- scales the adjacency matrix such that the features of neighboring nodes are weighted appropriately during
- aggregation. The steps involved in the renormalization trick are as follows:
- - Compute the degree matrix.
- - Compute the inverse square root of the degree matrix.
- - Multiply the inverse square root of the degree matrix with the adjacency matrix.
- """
- # Set the paths to the data files
- content_path = os.path.join(path, 'cora.content')
- cites_path = os.path.join(path, 'cora.cites')
- # Load data from files
- content_tensor = np.genfromtxt(content_path, dtype=np.dtype(str))
- cites_tensor = np.genfromtxt(cites_path, dtype=np.int32)
- # Process features
- features = torch.FloatTensor(content_tensor[:, 1:-1].astype(np.int32)) # Extract feature values
- scale_vector = torch.sum(features, dim=1) # Compute sum of features for each node
- scale_vector = 1 / scale_vector # Compute reciprocal of the sums
- scale_vector[scale_vector == float('inf')] = 0 # Handle division by zero cases
- scale_vector = torch.diag(scale_vector).to_sparse() # Convert the scale vector to a sparse diagonal matrix
- features = scale_vector @ features # Scale the features using the scale vector
- # Process labels
- classes, labels = np.unique(content_tensor[:, -1], return_inverse=True) # Extract unique classes and map labels to indices
- labels = torch.LongTensor(labels) # Convert labels to a tensor
- # Process adjacency matrix
- idx = content_tensor[:, 0].astype(np.int32) # Extract node indices
- idx_map = {id: pos for pos, id in enumerate(idx)} # Create a dictionary to map indices to positions
- # Map node indices to positions in the adjacency matrix
- edges = np.array(
- list(map(lambda edge: [idx_map[edge[0]], idx_map[edge[1]]],
- cites_tensor)), dtype=np.int32)
- V = len(idx) # Number of nodes
- E = edges.shape[0] # Number of edges
- adj_mat = torch.sparse_coo_tensor(edges.T, torch.ones(E), (V, V), dtype=torch.int64) # Create the initial adjacency matrix as a sparse tensor
- adj_mat = torch.eye(V) + adj_mat # Add self-loops to the adjacency matrix
- degree_mat = torch.sum(adj_mat, dim=1) # Compute the sum of each row in the adjacency matrix (degree matrix)
- degree_mat = torch.sqrt(1 / degree_mat) # Compute the reciprocal square root of the degrees
- degree_mat[degree_mat == float('inf')] = 0 # Handle division by zero cases
- degree_mat = torch.diag(degree_mat).to_sparse() # Convert the degree matrix to a sparse diagonal matrix
- adj_mat = degree_mat @ adj_mat @ degree_mat # Apply the renormalization trick
- return features.to_sparse().to(device), labels.to(device), adj_mat.to_sparse().to(device)
- def train_iter(epoch, model, optimizer, criterion, input, target, mask_train, mask_val, print_every=10):
- start_t = time.time()
- model.train()
- optimizer.zero_grad()
- # Forward pass
- output = model(*input)
- loss = criterion(output[mask_train], target[mask_train]) # Compute the loss using the training mask
- loss.backward()
- optimizer.step()
- # Evaluate the model performance on training and validation sets
- loss_train, acc_train = test(model, criterion, input, target, mask_train)
- loss_val, acc_val = test(model, criterion, input, target, mask_val)
- if epoch % print_every == 0:
- # Print the training progress at specified intervals
- print(f'Epoch: {epoch:04d} ({(time.time() - start_t):.4f}s) loss_train: {loss_train:.4f} acc_train: {acc_train:.4f} loss_val: {loss_val:.4f} acc_val: {acc_val:.4f}')
- def test(model, criterion, input, target, mask):
- model.eval()
- with torch.no_grad():
- output = model(*input)
- output, target = output[mask], target[mask]
- loss = criterion(output, target)
- acc = (output.argmax(dim=1) == target).float().sum() / len(target)
- return loss.item(), acc.item()
- if __name__ == '__main__':
- parser = argparse.ArgumentParser(description='PyTorch Graph Convolutional Network')
- parser.add_argument('--epochs', type=int, default=200,
- help='number of epochs to train (default: 200)')
- parser.add_argument('--lr', type=float, default=0.01,
- help='learning rate (default: 0.01)')
- parser.add_argument('--l2', type=float, default=5e-4,
- help='weight decay (default: 5e-4)')
- parser.add_argument('--dropout-p', type=float, default=0.5,
- help='dropout probability (default: 0.5)')
- parser.add_argument('--hidden-dim', type=int, default=16,
- help='dimension of the hidden representation (default: 16)')
- parser.add_argument('--val-every', type=int, default=20,
- help='epochs to wait for print training and validation evaluation (default: 20)')
- parser.add_argument('--include-bias', action='store_true',
- help='use bias term in convolutions (default: False)')
- parser.add_argument('--no-accel', action='store_true',
- help='disables accelerator')
- parser.add_argument('--dry-run', action='store_true',
- help='quickly check a single pass')
- parser.add_argument('--seed', type=int, default=42, metavar='S',
- help='random seed (default: 42)')
- args = parser.parse_args()
- use_accel = not args.no_accel and torch.accelerator.is_available()
- torch.manual_seed(args.seed)
- if use_accel:
- device = torch.accelerator.current_accelerator()
- else:
- device = torch.device('cpu')
- print(f'Using {device} device')
- cora_url = 'https://linqs-data.soe.ucsc.edu/public/lbc/cora.tgz'
- print('Downloading dataset...')
- with requests.get(cora_url, stream=True) as tgz_file:
- with tarfile.open(fileobj=tgz_file.raw, mode='r:gz') as tgz_object:
- tgz_object.extractall()
- print('Loading dataset...')
- features, labels, adj_mat = load_cora(device=device)
- idx = torch.randperm(len(labels)).to(device)
- idx_test, idx_val, idx_train = idx[:1000], idx[1000:1500], idx[1500:]
- gcn = GCN(features.shape[1], args.hidden_dim, labels.max().item() + 1, args.include_bias, args.dropout_p).to(device)
- optimizer = Adam(gcn.parameters(), lr=args.lr, weight_decay=args.l2)
- criterion = nn.NLLLoss()
- for epoch in range(args.epochs):
- train_iter(epoch + 1, gcn, optimizer, criterion, (features, adj_mat), labels, idx_train, idx_val, args.val_every)
- if args.dry_run:
- break
- loss_test, acc_test = test(gcn, criterion, (features, adj_mat), labels, idx_test)
- print(f'Test set results: loss {loss_test:.4f} accuracy {acc_test:.4f}')
main.py at commit acc295d, under BSD-3-Clause · at the source
Overview
- Department of Information Science, Faculty of Science, Toho University,2-2-1 Miyama, Funabashi, Chiba, 274-8510 Japan
- School of Science and Engineering, Tokyo Denki University,Saitama, Japan
Abstract
The abstract is not reproduced here: the paper's license (CC BY-NC-ND) does not allow it. Read it in the paper, at the publisher or on Europe PMC.
Repositories
Its files are read in the Code ↔ Paper reader above, with 3 matches between paragraphs and lines of code.
pytorch/examples
acc295dc7b90714f1bf47f06004fc19a7fe235c4, 1 September 2025Availability: 1 check, the latest on 28 September 2026: the link answers
- 28 September 2026: the link answers
117 files
- cpp/
autograd/ , C++, 191 linesautograd.cpp - cpp/
custom-dataset/ , C++, 231 linescustom-dataset.cpp - cpp/
dcgan/ , C++, 217 linesdcgan.cpp - cpp/
dcgan/ , Python, 28 linesdisplay_samples.py - cpp/
distributed/ , C++, 185 linesdist-mnist.cpp - cpp/
mnist/ , C++, 154 linesmnist.cpp - cpp/
regression/ , C++, 91 linesregression.cpp - cpp/
tools/ , Python, 89 linesdownload_mnist.py - cpp/
transfer-learning/ , C++, 78 linesclassify.cpp - cpp/
transfer-learning/ , Python, 19 linesconvert.py - cpp/
transfer-learning/ , C++, 266 linesmain.cpp - cpp/
transfer-learning/ , C/C++, 68 linesmain.h - dcgan/
main.py , Python, 277 lines - distributed/
FSDP/ , Python, 217 linesT5_training.py - distributed/
FSDP/ , Python, 2 linesconfigs/ __init__.py - distributed/
FSDP/ , Python, 19 linesconfigs/ fsdp.py - distributed/
FSDP/ , Python, 19 linesconfigs/ training.py - distributed/
FSDP/ , Shell, 8 linesdownload_dataset.sh - distributed/
FSDP/ , Python, 10 linesmodel_checkpointing/ __init__.py - distributed/
FSDP/ , Python, 307 linesmodel_checkpointing/ checkpoint_handler.py - distributed/
FSDP/ , Python, 3 linespolicies/ __init__.py - distributed/
FSDP/ , Python, 31 linespolicies/ activation_checkpointing _functions.py - distributed/
FSDP/ , Python, 38 linespolicies/ mixed_precision.py - distributed/
FSDP/ , Python, 47 linespolicies/ wrapping.py - distributed/
FSDP/ , Python, 82 linessummarization_dataset.py - distributed/
FSDP/ , Python, 4 linesutils/ __init__.py - distributed/
FSDP/ , Python, 27 linesutils/ environment.py - distributed/
FSDP/ , Python, 102 linesutils/ train_utils.py - distributed/
FSDP2/ , Python, 209 linescheckpoint.py - distributed/
FSDP2/ , Python, 118 linesexample.py - distributed/
FSDP2/ , Python, 134 linesmodel.py - distributed/
FSDP2/ , Shell, 11 linesrun_example.sh - distributed/
FSDP2/ , Python, 24 linesutils.py - distributed/
ddp-tutorial-series/ , Python, 13 linesdatautils.py - distributed/
ddp-tutorial-series/ , Python, 104 linesmultigpu.py - distributed/
ddp-tutorial-series/ , Python, 111 linesmultigpu_torchrun.py - distributed/
ddp-tutorial-series/ , Python, 112 linesmultinode.py - distributed/
ddp-tutorial-series/ , Python, 82 linessingle_gpu.py - distributed/
ddp-tutorial-series/ , Shell, 23 linesslurm/ sbatch_run.sh - distributed/
ddp/ , Python, 101 linesexample.py - distributed/
ddp/ , Shell, 10 linesrun_example.sh - distributed/
minGPT-ddp/ , Python, 43 linesmingpt/ char_dataset.py - distributed/
minGPT-ddp/ , Python, 57 linesmingpt/ main.py - distributed/
minGPT-ddp/ , Python, 252 linesmingpt/ model.py - distributed/
minGPT-ddp/ , Shell, 25 linesmingpt/ slurm/ sbatch_run.sh - distributed/
minGPT-ddp/ , Python, 154 linesmingpt/ trainer.py - distributed/
minGPT-ddp/ , Shell, 10 linesrun_example.sh - distributed/
rpc/ , Python, 158 linesbatch/ parameter_server.py - distributed/
rpc/ , Python, 262 linesbatch/ reinforce.py - distributed/
rpc/ , Python, 198 linesddp_rpc/ main.py - distributed/
rpc/ , Python, 312 linesparameter_server/ rpc_parameter_server.py - distributed/
rpc/ , Python, 288 linespipeline/ main.py - distributed/
rpc/ , Python, 250 linesrl/ main.py - distributed/
rpc/ , Python, 86 linesrnn/ main.py - distributed/
rpc/ , Python, 106 linesrnn/ rnn.py - distributed/
tensor_parallelism/ , Python, 178 linesfsdp_tp_example.py - distributed/
tensor_parallelism/ , Python, 462 linesllama2_model.py - distributed/
tensor_parallelism/ , Python, 22 lineslog_utils.py - distributed/
tensor_parallelism/ , Shell, 13 linesrun_example.sh - distributed/
tensor_parallelism/ , Python, 113 linessequence_parallel_exampl e.py - distributed/
tensor_parallelism/ , Python, 127 linestensor_parallel_example. py - docs/
source/ , Python, 76 linesconf.py - fast_neural_style/
download_saved_models.py , Python, 28 lines - fast_neural_style/
neural_style/ , Python, 1 line__init__.py - fast_neural_style/
neural_style/ , Python, 261 linesneural_style.py - fast_neural_style/
neural_style/ , Python, 99 linestransformer_net.py - fast_neural_style/
neural_style/ , Python, 34 linesutils.py - fast_neural_style/
neural_style/ , Python, 38 linesvgg.py - fx/
custom_tracer.py , Python, 116 lines - fx/
inline_function.py , Python, 70 lines - fx/
invert.py , Python, 62 lines - fx/
module_tracer.py , Python, 148 lines - fx/
native_interpreter/ , C++, 179 linesinterpreter.cpp - fx/
native_interpreter/ , Python, 129 linesuse_interpreter.py - fx/
primitive_library.py , Python, 166 lines - fx/
profiling_tracer.py , Python, 119 lines - fx/
proxy_based_graph_creati , Python, 65 lineson.py - fx/
replace_op.py , Python, 63 lines - fx/
subgraph_rewriter_basic_ , Python, 72 linesuse.py - fx/
wrap_output_dynamically. , Python, 85 linespy - gat/
main.py , Python, 371 lines - gcn/
main.py , Python, 256 lines, 1 match - imagenet/
extract_ILSVRC.sh , Shell, 80 lines - imagenet/
main.py , Python, 530 lines, 1 match - language_translation/
main.py , Python, 306 lines - language_translation/
src/ , Python, 134 linesdata.py - language_translation/
src/ , Python, 98 linesmodel.py - legacy/
snli/ , Python, 76 linesmodel.py - legacy/
snli/ , Python, 146 linestrain.py - legacy/
snli/ , Python, 71 linesutil.py - mnist/
main.py , Python, 141 lines - mnist_forward_forward/
main.py , Python, 174 lines - mnist_hogwild/
main.py , Python, 105 lines - mnist_hogwild/
train.py , Python, 56 lines - mnist_rnn/
main.py , Python, 142 lines - regression/
main.py , Python, 68 lines - reinforcement_learning/
actor_critic.py , Python, 181 lines - reinforcement_learning/
reinforce.py , Python, 107 lines - run_cpp_examples.sh, Shell, 183 lines
- run_distributed_examples
.sh , Shell, 101 lines - run_python_examples.sh, Shell, 247 lines
- siamese_network/
main.py , Python, 302 lines - super_resolution/
data.py , Python, 70 lines - super_resolution/
dataset.py , Python, 37 lines - super_resolution/
main.py , Python, 84 lines - super_resolution/
model.py , Python, 30 lines - super_resolution/
super_resolve.py , Python, 52 lines - time_sequence_prediction
/ , Python, 13 linesgenerate_sine_wave.py - time_sequence_prediction
/ , Python, 90 linestrain.py - utils.sh, Shell, 55 lines
- vae/
main.py , Python, 139 lines - word_language_model/
data.py , Python, 48 lines - word_language_model/
generate.py , Python, 96 lines - word_language_model/
main.py , Python, 290 lines - word_language_model/
model.py , Python, 144 lines - LICENSE, License, 29 lines
- README.md, Text, 39 lines
naoto0804/pytorch-AdaIN
47950d0e6656a95a80a4b105c4c0f58d38ef785c, 23 January 2024Availability: 1 check, the latest on 28 September 2026: the link answers
- 28 September 2026: the link answers
9 files
- function.py, Python, 67 lines
- net.py, Python, 152 lines
- sampler.py, Python, 26 lines
- test.py, Python, 161 lines
- test_video.py, Python, 200 lines
- torch_to_pytorch.py, Python, 322 lines, 1 match
- train.py, Python, 136 lines
- LICENSE, License, 21 lines
- README.md, Text, 61 lines
The paper's code and data availability statement is in the Data section.
Tracing map
Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.
What the map holds:
- 2 repositories of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
- 122 scripts, each with its path and the digest of its content;
- 3 matches between paragraphs of the paper and lines of the code (method lexical-v1);
- neither the text of the paper nor the code itself.
Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.
Data
No dataset and no data link were found in the paper.
Code and data availability statement
The paper has a code and data availability statement. Its license (CC BY-NC-ND) does not allow reproducing it here; in short, from what the harvester recognized in it:
- it points to the authors' code: pytorch/
examples - it says that the data are available on request
Read it in the paper: doi.org/10.1038/s41598-026-53152-9.
Versions
The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.
Version 1, 28 September 2026: the first record
Recorded: type, language, journal, volume, issue, pages, dates, 3 authors, 7 keywords, 2 funders, 42 references.
Cite
This paper
Wagatsuma, N., Ito, K., & Hidaka, A. (2026). Layer-specific feature preference in a trained AlexNet model revealed by stylized image analysis. Scientific reports, 16(1), 22791. https://
BibTeX
@article{wagatsuma2026la
author = {Wagatsuma, Nobuhiko and Ito, Kazuma and Hidaka, Akinori},
title = {{Layer-specific feature preference in a trained AlexNet model revealed by stylized image analysis}},
journal = {Scientific reports},
year = {2026},
month = may,
volume = {16},
number = {1},
pages = {22791},
publisher = {Nature Publishing Group},
issn = {2045-2322},
doi = {10.1038/
url = {https://
pmid = {42156935},
pmcid = {PMC13385648}
}
RIS
TY - JOUR
AU - Wagatsuma, Nobuhiko
AU - Ito, Kazuma
AU - Hidaka, Akinori
TI - Layer-specific feature preference in a trained AlexNet model revealed by stylized image analysis
T2 - Scientific reports
J2 - Sci Rep
PY - 2026
DA - 2026/
VL - 16
IS - 1
SP - 22791
SN - 2045-2322
PB - Nature Publishing Group
DO - 10.1038/
UR - https://
LA - en
ER -
CSL-JSON
{
"id": "10.1038/
"type": "article-journal",
"title": "Layer-specific feature preference in a trained AlexNet model revealed by stylized image analysis",
"container-title": "Scientific reports",
"author": [
{
"family": "Wagatsuma",
"given": "Nobuhiko"
},
{
"family": "Ito",
"given": "Kazuma"
},
{
"family": "Hidaka",
"given": "Akinori"
}
],
"container-title-short":
"volume": "16",
"issue": "1",
"page": "22791",
"DOI": "10.1038/
"PMID": "42156935",
"PMCID": "PMC13385648",
"ISSN": "2045-2322",
"publisher": "Nature Publishing Group",
"URL": "https://
"language": "en",
"issued": {
"date-parts": [
[
2026,
5,
19
]
]
}
}
The tracing map gets a citation of its own once an author has validated it and it has a DOI.
Similar papers
The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.
- [1] doi:10.1016/j.isci.2025.112199 [code]
- Brain-like border ownership signals support prediction of natural videos.Journal: iScienceIn common: imageio, OpenCV, Pillow, 3 other tools, 4 references
- [2] doi:10.1162/opmi_a_00189 [code]
- Approximating Human-Level 3D Visual Inferences With Deep Neural Networks.Journal: Open mind : discoveries in cognitive scienceIn common: imageio, OpenCV, Pillow, 4 other tools, 4 references
- [3] doi:10.1038/s42003-025-09411-y [code]
- Structured interconnectivity optimizes neural geometry for balancing specificity and generalization in object recognition.Journal: Communications biologyIn common: imageio, Pillow, PyTorch, 3 other tools, 5 references
- [4] doi:10.1167/jov.22.7.3 [code]
- Could simplified stimuli change how the brain performs visual search tasks? A deep neural network study.Journal: Journal of visionIn common: imageio, OpenCV, Pillow, 4 other tools, 4 references
- [5] doi:10.1038/s41467-024-45679-0 [code]
- Improved modeling of human vision by incorporating robustness to blur in convolutional neural networks.Journal: Nature communicationsIn common: OpenCV, Pillow, PyTorch, 3 other tools, 5 references
- [6] doi:10.1126/sciadv.abm2219 [code]
- Reassessing hierarchical correspondences between brain and deep networks through direct interface.Journal: Science advancesIn common: Hugging Face Transformers, imageio, OpenCV, 5 other tools, 2 references
- [7] doi:10.1038/s41562-025-02252-z [code]
- In silico discovery of representational relationships across visual cortex.Journal: Nature human behaviourIn common: Pillow, PyTorch, pandas, 2 other tools, 6 references
- [8] doi:10.1371/journal.pcbi.1013391 [code]
- A feedforward mechanism for human-like contour integration.Journal: PLoS computational biologyIn common: OpenCV, Pillow, PyTorch, 3 other tools, 5 references
- [9] doi:10.1038/s42003-026-09685-w [code]
- Achieving more human brain-like vision via human EEG representational alignment.Journal: Communications biologyIn common: Hugging Face Transformers, imageio, OpenCV, 5 other tools, 2 references
- [10] doi:10.1126/sciadv.adi1480 [code]
- SpikingJelly: An open-source machine learning infrastructure platform for spike-based intelligence.Journal: Science advancesIn common: Hugging Face Transformers, OpenCV, Pillow, 4 other tools, 3 references
Contribute
The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.
Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.
Claim this paper
Correct its record
Say what each link of this record is, remove the ones that are not the paper's, add the ones that are missing. The correction becomes a new version of the record, in its Versions section.
Validate its tracing map
You validate the map as this page shows it: 2 repositories of the authors' code, each at its verified commit and with its license, 122 scripts, and 3 matches between paragraphs and code (see the Code and Map sections). It then receives a DOI on Zenodo, with you (your ORCID iD) and OSCR as its creators; the code itself is not deposited.
The map's fingerprint: sha256:8b71d3ff873a6087…
Add the badge to the code's README
If you are an author or a maintainer of this paper's code, you can add this badge. It says what OSCR knows of the paper (its code, its validated tracing map) and links the code to this page. Paste one of these snippets under the README's title. Keep the two marks around it, so that the badge can be replaced in place and is never added twice. OSCR changes nothing in your repository.
Markdown
<!-- oscr-badge:start -->
[](https://openscicode.org/badge/doi/10.1038/s41598-026-53152-9)
<!-- oscr-badge:end -->reStructuredText
.. oscr-badge:start
.. image:: https://openscicode.org/badge/doi/10.1038/s41598-026-53152-9.svg
:target: https://openscicode.org/badge/doi/10.1038/s41598-026-53152-9
:alt: OSCR
.. oscr-badge:endHTML
<!-- oscr-badge:start -->
<a href="https://openscicode.org/badge/doi/10.1038/s41598-026-53152-9"><img src="https://openscicode.org/badge/doi/10.1038/s41598-026-53152-9.svg" alt="OSCR"></a>
<!-- oscr-badge:end -->On GitHub, open the README, choose Edit (the pencil), paste the snippet under the title, then "Commit changes".
If you choose to, you can have the badge kept up to date for you, with oscr badge --write from the command line or with the GitHub Action in your repository's workflows. Every other form and option (a card, AsciiDoc, LaTeX, shields.io) is on the badge page.
Request its removal
To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).
Discussion, reproductions, activity
Discussion: code errors and mismatches between the code and the paper about this paper and its code, from readers and its authors, beside the paper and its tracing map.
Report a mismatch between the code and the paper · Report a code error (signed in)
Reproductions: reports from readers who ran the authors' code and did not get the paper's result, what they ran, with which environment, commit and data.
Report a reproduction failure (signed in)
Versions of its code: the releases of its code tied to a version of this paper (the preprint, the accepted manuscript, the version of record…), each with the tracing map versioned with it.
Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.
