SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1512.00567 · CVPR · 2015

Rethinking the Inception Architecture for Computer Vision

Christian Szegedy, Zbigniew Wojna

arXiv · PDF · Open in the Atlas

Code that ran

20 of 26 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; 3 honoured the contract we drafted, 1 violated it and 12 have no contract verdict; 4 failures of our instrument; 6 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/1512.00567.json · badge: /api/ran/1512.00567.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
matthias-wright/jax-fid community (classifier label)
classifier basis not recorded on the link
4 ran · 2 no recorded run
0jason000/inception_v3 community (classifier label)
classifier basis not recorded on the link
4 ran · 1 no recorded run
hwzwx1190522/inceptionv3 community (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
athulyashyju/sample-project community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
KyleYoung1997/DNNorDermatologist community (classifier label)
classifier basis not recorded on the link
2 ran · 1 no recorded run
calmisential/TensorFlow2.0_InceptionV3 community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
LKLQQ/Inception-v2 community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
mjhucla/TF-mRNN unrelated (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
DanishShah/DeepDiagnosis community (classifier label)
classifier basis not recorded on the link
0 ran · 2 no recorded run
FunctionStatusWhere it lives
compute_frechet_distance Ran
ran on a synthesized input
matthias-wright/jax-fid/jax_fid/fid.py
code served (permissive licence) · get_code("ab8a4e89eb6d87ad")
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
import scipy

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

# Create small but distinct-sized covariance matrices and mean vectors
# Using dimension 4 for the feature space
dim = 4

# Create two mean vectors
mu1 = np.random.randn(dim)
mu2 = np.random.randn(dim)

# Create two positive semi-definite covariance matrices
# Using random matrices and their transpose to ensure PSD property
A1 = np.random.randn(dim, dim)
sigma1 = A1 @ A1.T + np.eye(dim) * 0.1

A2 = np.random.randn(dim, dim)
sigma2 = A2 @ A2.T + np.eye(dim) * 0.1

ARGS = {
    'mu1': mu1,
    'mu2': mu2,
    'sigma1': sigma1,
    'sigma2': sigma2,
    '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
compute_statistics Ran
ran on a synthesized input
matthias-wright/jax-fid/jax_fid/fid.py
code served (permissive licence) · get_code("a5a4622777935741")
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 jax
import jax.numpy as jnp
import numpy as np
from PIL import Image
import os
import tempfile
import shutil

# Create a temporary directory with test images
temp_dir = tempfile.mkdtemp()

# Create a few small test images
for i in range(3):
    img_array = np.random.randint(0, 256, (32, 32, 3), dtype=np.uint8)
    img = Image.fromarray(img_array, 'RGB')
    img.save(os.path.join(temp_dir, f'test_img_{i}.png'))

# Define a minimal apply_fn that works with JAX
def apply_fn(params, x):
    # Simple function that processes input and returns activations
    # x shape: (batch_size, height, width, channels)
    # Output should have spatial dimensions that can be squeezed
    batch_size = x.shape[0]
    # Return shape (batch_size, 1, 1, 4) so squeezing gives (batch_size, 4)
    return jnp.ones((batch_size, 1, 1, 4))

# Create minimal params (can be empty dict or None-like structure)
params = {}

ARGS = {
    'path': temp_dir,
    'params': params,
    'apply_fn': apply_fn,
    'batch_size': 2,
    'img_size': (32, 32)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, jax, numpy, tqdm — 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
create_image_lists Ran
ran on a synthesized input
athulyashyju/sample-project/train_data.py
pointer only (licence: Apache-2.0) · get_code("8fcfb76b09017d43")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, athulyashyju/sample-project, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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.

fill_segmentation Ran
ran on a synthesized input
KyleYoung1997/DNNorDermatologist/Shap_GradCAM_Notebooks/utils.py
code served (permissive licence) · get_code("c92793f2c1418164")
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 2, dtype family float, shape (3, 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 numpy as np

# Construct arguments for fill_segmentation function
# The function takes:
# 1. values: array-like of values to fill with
# 2. segmentation: array with integer indices that determine which cells get which values

# Create a simple segmentation array with indices 0, 1, 2
segmentation = np.array([
    [0, 0, 1, 1],
    [0, 0, 1, 1],
    [2, 2, 2, 2]
])

# Create values array with one value for each segmentation index (0, 1, 2)
values = np.array([10.0, 20.0, 30.0])

ARGS = {
    'values': values,
    'segmentation': segmentation
}
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 2, dtype family float, shape (3, 4) — shape (3, 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
float_parameter Ran
ran, honoured the contract we drafted
0jason000/inception_v3/src/transforms.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from chenchenzong/dpc, is pointer only (licence: NONE) · get_code("40d3c39a403ae1f7")
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 chenchenzong/dpc, 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 Ran
ran; our contract draft was wrong, not the code
matthias-wright/jax-fid/jax_fid/utils.py
code served (permissive licence) · get_code("c98a6cf41bf8d8ac")
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 str. 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

torch.manual_seed(0)

# The function get(dictionary, key) takes:
# - dictionary: a dict or None
# - key: a hashable key to look up in the dictionary

# We need to construct valid arguments for both parameters
ARGS = {
    "dictionary": {"key1": "value1", "key2": 42, "key3": [1, 2, 3]},
    "key": "key1"
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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_bottleneck_path Ran
ran on a synthesized input
athulyashyju/sample-project/train_data.py
pointer only (licence: Apache-2.0) · get_code("6edc2cf4b05e4213")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, athulyashyju/sample-project, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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_image_path Ran
ran on a synthesized input
athulyashyju/sample-project/train_data.py
pointer only (licence: Apache-2.0) · get_code("b197ee6e37fc8268")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, athulyashyju/sample-project, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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_images_and_labels Ran
ran on a synthesized input
calmisential/TensorFlow2.0_InceptionV3/prepare_data.py
code served (permissive licence) · get_code("78374ffdcfcb124c")
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 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 pathlib
import tempfile
import os

# Create a temporary directory structure with subdirectories (labels) and dummy image files
temp_dir = tempfile.mkdtemp()
label_dirs = ['cat', 'dog']
for label in label_dirs:
    label_path = os.path.join(temp_dir, label)
    os.makedirs(label_path, exist_ok=True)
    # Create dummy image files
    for i in range(2):
        image_file = os.path.join(label_path, f'image_{i}.jpg')
        open(image_file, 'w').close()

ARGS = {
    'data_root_dir': temp_dir
}
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 tuple, ndim 2, 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_lr Ran
ran on a synthesized input
LKLQQ/Inception-v2/src/lr_generator.py
code served (permissive licence) · get_code("36cd7310bc74d09d")
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 float, shape (1000,) 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 math
import numpy as np

# Define ARGS for the get_lr function
ARGS = {
    'lr_init': 0.0001,
    'lr_end': 0.00001,
    'lr_max': 0.001,
    'warmup_epochs': 2,
    'total_epochs': 10,
    'steps_per_epoch': 100,
    'lr_decay_mode': 'cosine',
    'global_step': 0
}
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 float, shape (1000,) — shape (1000,) 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
init_group_params Ran
ran on a synthesized input
0jason000/inception_v3/src/optim.py
code served (permissive licence) · get_code("698131638ff8d4ad")
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 list, 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
import torch.nn as nn

torch.manual_seed(0)

class Param:
    def __init__(self, name):
        self.name = name
        self.data = torch.randn(2, 4)

params = [
    Param('weight'),
    Param('bias'),
    Param('beta'),
    Param('gamma'),
    Param('kernel'),
    Param('scale')
]

weight_decay = 0.01

ARGS = {
    'params': params,
    'weight_decay': weight_decay
}
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 list, ndim 1, dtype family float — a list 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
int_parameter Ran
ran, honoured the contract we drafted
0jason000/inception_v3/src/transforms.py
code served (permissive licence) · get_code("e5a370ee6856bd8d")
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 int, ndim 0, dtype family int, 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 torch
import numpy as np

ARGS = {
    "level": 5,
    "maxval": 100
}
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 int, ndim 0, dtype family int, 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
is_model Ran
ran, violated the contract we drafted
0jason000/inception_v3/src/model.py
code served (permissive licence) · get_code("b85a6569fd4f3890")
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 bool, 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 torch

# Set up the global model entrypoints dictionary with some test models
_model_entrypoints = {
    'resnet18': 'dummy_model_1',
    'vit_base': 'dummy_model_2',
    'bert_small': 'dummy_model_3'
}

# Define arguments for is_model function
ARGS = {
    'model_name': 'resnet18'
}
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 bool, 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
mask_image Ran
ran on a synthesized input
KyleYoung1997/DNNorDermatologist/Shap_GradCAM_Notebooks/utils.py
code served (permissive licence) · get_code("c9008a85f4e8d475")
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 4, dtype family float, shape (2, 4, 4, 3) 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 small test arrays
np.random.seed(0)

# zs: shape (batch_size, num_segments) - binary mask indicating which segments to keep
# Values should be 0 or 1
zs = np.array([[1, 0, 1, 0],
               [0, 1, 1, 0]], dtype=np.int32)

# segmentation: shape (height, width) - integer labels for each pixel
# Should have values from 0 to num_segments-1
segmentation = np.array([[0, 1, 2, 3],
                         [1, 2, 3, 0],
                         [2, 3, 0, 1],
                         [3, 0, 1, 2]], dtype=np.int32)

# image: shape (height, width, channels) - the image to mask
image = np.random.randn(4, 4, 3).astype(np.float32)

# background: shape (channels,) or None - the replacement color
# If None, will be computed as mean of image along spatial dims
background = None

ARGS = {
    "zs": zs,
    "segmentation": segmentation,
    "image": image,
    "background": background
}
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 4, dtype family float, shape (2, 4, 4, 3) — shape (2, 4, 4, 3) 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
merge Ran
ran; our contract draft was wrong, not the code
hwzwx1190522/inceptionv3/inceptionv3/src/model_utils/config.py
code served (permissive licence) · get_code("b9e24c3e3e0e88a7")
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) 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 argparse

torch.manual_seed(0)

# Create a mock args object with some attributes
args = argparse.Namespace(
    learning_rate=0.001,
    batch_size=32,
    epochs=10,
    model_name="test_model",
    debug=True,
    seed=42
)

# Create a base configuration dictionary
cfg = {
    "default_lr": 0.0001,
    "default_batch": 16,
    "dataset": "cifar10"
}

ARGS = {
    "args": args,
    "cfg": cfg
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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
parse_cli_to_yaml Ran
ran; our contract draft was wrong, not the code
hwzwx1190522/inceptionv3/inceptionv3/src/model_utils/config.py
code served (permissive licence) · get_code("738456e1b7388e7e")
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 Namespace. 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 argparse
import ast

# Create minimal parent parser
parent_parser = argparse.ArgumentParser(add_help=False)

# Create base configuration dict with various types
cfg = {
    'learning_rate': 0.001,
    'batch_size': 32,
    'epochs': 10,
    'use_cuda': False,
    'model_name': 'resnet50',
    'dropout': 0.5,
    'debug': True,
    'nested_list': [1, 2, 3],  # This will be skipped since it's a list
    'nested_dict': {'a': 1},   # This will be skipped since it's a dict
}

# Create helper descriptions
helper = {
    'learning_rate': 'Learning rate for optimizer',
    'batch_size': 'Batch size for training',
    'epochs': 'Number of training epochs',
    'use_cuda': 'Whether to use CUDA',
    'model_name': 'Name of the model architecture',
    'dropout': 'Dropout rate',
    'debug': 'Enable debug mode',
}

# Create choices for specific parameters
choices = {
    'model_name': ['resnet50', 'vgg16', 'mobilenet'],
}

# Set config path
cfg_path = 'default_config.yaml'

ARGS = {
    'parser': parent_parser,
    'cfg': cfg,
    'helper': helper,
    'choices': choices,
    'cfg_path': cfg_path,
}
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 Namespace — a Namespace output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Namespace 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
parse_yaml Ran
ran on a synthesized input
hwzwx1190522/inceptionv3/inceptionv3/src/model_utils/config.py
code served (permissive licence) · get_code("96a9939953d5992d")
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 tempfile
import os
import yaml

# Create a temporary YAML file with valid config structure
temp_dir = tempfile.mkdtemp()
yaml_path = os.path.join(temp_dir, "config.yaml")

# Write a valid YAML file with 3 documents: config, helper, and choices
yaml_content = """
model: resnet50
learning_rate: 0.001
batch_size: 32
---
model: "Model architecture to use"
learning_rate: "Learning rate for optimization"
batch_size: "Batch size for training"
---
model: [resnet50, resnet101, vgg16]
learning_rate: [0.0001, 0.001, 0.01]
batch_size: [16, 32, 64]
"""

with open(yaml_path, 'w') as f:
    f.write(yaml_content)

ARGS = {
    "yaml_path": yaml_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
yaml — 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
pool Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
matthias-wright/jax-fid/jax_fid/inception.py
code served (permissive licence) · get_code("19dd078b5186d077")
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 ArrayImpl, ndim 4, dtype family float, shape (2, 2, 1, 1) 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 jax
import jax.numpy as jnp

# Set seed for reproducibility
import numpy as np
np.random.seed(0)

# Create input tensor: (batch=2, height=4, width=3, channels=1)
inputs = jnp.ones((2, 4, 3, 1))

# Initial value for reduction (using 0 for max pooling with -inf would be better, but 0 works for sum)
init = -jnp.inf

# Define a simple reduce function (max operation)
reduce_fn = jax.lax.max

# Window shape for 2D pooling: (2, 2)
window_shape = (2, 2)

# Strides for the pooling operation
strides = (2, 2)

# Padding specification: 'VALID' means no padding
padding = 'VALID'

ARGS = {
    'inputs': inputs,
    'init': init,
    'reduce_fn': reduce_fn,
    'window_shape': window_shape,
    'strides': strides,
    'padding': padding
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 4, dtype family float, shape (2, 2, 1, 1) — shape (2, 2, 1, 1) 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
read_label Ran
ran on a synthesized input
hwzwx1190522/inceptionv3/inceptionv3/postprocess.py
code served (permissive licence) · get_code("14f5f199a7b894be")
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 tempfile
import os

# Create a temporary label file with sample data
temp_dir = tempfile.mkdtemp()
label_file = os.path.join(temp_dir, "labels.txt")

# Write sample label data
with open(label_file, 'w') as f:
    f.write("img_001:cat\n")
    f.write("img_002:dog\n")
    f.write("img_003:bird\n")

ARGS = {
    "label_file": label_file
}
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 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
run_epoch Ran
ran, honoured the contract we drafted
mjhucla/TF-mRNN/exp/ms_coco_caption/mrnn_trainer_mscoco.py
pointer only (licence: NONE) · get_code("0e80fb39f95dde2a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, mjhucla/TF-mRNN 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.

avg_pool Not yet run
unverified (harvested, no recorded run)
matthias-wright/jax-fid/jax_fid/inception.py
code served (permissive licence) · get_code("9163602355a43f00")
How it ran

no invocation recorded.

classify Not yet run
unverified (harvested, no recorded run)
DanishShah/DeepDiagnosis/Web/Backend/predict.py
code served (permissive licence) · get_code("1d804546ecbde837")
How it ran

no invocation recorded.

classify_retino Not yet run
unverified (harvested, no recorded run)
DanishShah/DeepDiagnosis/Web/Backend/predict.py
code served (permissive licence) · get_code("de83097dace9ff51")
How it ran

no invocation recorded.

download Not yet run
unverified (harvested, no recorded run)
matthias-wright/jax-fid/jax_fid/utils.py
code served (permissive licence) · get_code("f93ac625244c51c4")
How it ran

no invocation recorded.

make_model Not yet run
unverified (harvested, no recorded run)
KyleYoung1997/DNNorDermatologist/DataSplit_HpSearch.py
code served (permissive licence) · get_code("4b446aabd3bcf05f")
How it ran

no invocation recorded.

model_entrypoint Not yet run
unverified (harvested, no recorded run)
0jason000/inception_v3/src/model.py
code served (permissive licence) · get_code("ac95eb88d9b993fd")
How it ran

no invocation recorded.

Also present in other papers' code

Of 26 samples lifted from this paper's linked repositories, 26 carry a function identity and 0 do not. Of the 26 distinct functions with an identity, 12 are also present in other papers whose code we hold (9 with at least one paper dated earlier, 3 the same month, 11 later) and 14 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
merge 28 other papers
2 dated earlier, 1 the same month, 25 later
1406.1078 (2014-06)
Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
1409.3215 (2014-09)
Sequence to Sequence Learning with Neural Networks
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition
and 25 more not named here
is_model 22 other papers
1 dated earlier, 21 later
1409.4842 (2014-09)
Going deeper with convolutions
1608.06993 (2016-08)
Densely Connected Convolutional Networks
1911.11907 (2019-11)
GhostNet: More Features from Cheap Operations
and 19 more not named here
parse_cli_to_yaml 20 other papers
2 dated earlier, 1 the same month, 17 later
1406.1078 (2014-06)
Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
1409.3215 (2014-09)
Sequence to Sequence Learning with Neural Networks
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition
and 17 more not named here
float_parameter 15 other papers
1 dated earlier, 14 later
1409.4842 (2014-09)
Going deeper with convolutions
1608.06993 (2016-08)
Densely Connected Convolutional Networks
1904.12848 (2019-04)
Unsupervised Data Augmentation for Consistency Training
and 12 more not named here
parse_yaml 15 other papers
2 dated earlier, 1 the same month, 12 later
1406.1078 (2014-06)
Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
1409.3215 (2014-09)
Sequence to Sequence Learning with Neural Networks
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition
and 12 more not named here
int_parameter 8 other papers
1 dated earlier, 7 later
1409.4842 (2014-09)
Going deeper with convolutions
1608.06993 (2016-08)
Densely Connected Convolutional Networks
1911.11907 (2019-11)
GhostNet: More Features from Cheap Operations
and 5 more not named here
init_group_params 5 other papers
1 dated earlier, 4 later
1409.4842 (2014-09)
Going deeper with convolutions
1608.06993 (2016-08)
Densely Connected Convolutional Networks
1911.11907 (2019-11)
GhostNet: More Features from Cheap Operations
and 2 more not named here
model_entrypoint 5 other papers
1 dated earlier, 4 later
1409.4842 (2014-09)
Going deeper with convolutions
1608.06993 (2016-08)
Densely Connected Convolutional Networks
1911.11907 (2019-11)
GhostNet: More Features from Cheap Operations
and 2 more not named here
download 1 other paper
1 later
2311.15100 (2023-11)
Unbalancedness in Neural Monge Maps Improves Unpaired Domain Translation
get 1 other paper
1 later
2311.15100 (2023-11)
Unbalancedness in Neural Monge Maps Improves Unpaired Domain Translation
pool 1 other paper
1 later
2311.15100 (2023-11)
Unbalancedness in Neural Monge Maps Improves Unpaired Domain Translation
run_epoch 1 other paper
1 dated earlier
1412.6632 (2014-12)
Deep Captioning with Multimodal Recurrent Neural Networks (m-RNN)

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

More are linked than shown here.

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

Abstract

Convolutional networks are at the core of most stateof-the-art computer vision solutions for a wide variety of tasks. Since 2014 very deep convolutional networks started to become mainstream, yielding substantial gains in various benchmarks. Although increased model size and computational cost tend to translate to immediate quality gains for most tasks (as long as enough labeled data is provided for training), computational efficiency and low parameter count are still enabling factors for various use cases such as mobile vision and big-data scenarios. Here we are exploring ways to scale up networks in ways that aim at utilizing the added computation as efficiently as possible by suitably factorized convolutions and aggressive regularization. We benchmark our methods on the ILSVRC 2012 classification challenge validation set demonstrate substantial gains over the state of the art: 21.2% top-1 and 5.6% top-5 error for single frame evaluation using a network with a computational cost of 5 billion multiply-adds per inference and with using less than 25 million parameters. With an ensemble of 4 models and multi-crop evaluation, we report 3.5% top-5 error and 17.3% top-1 error.

For agents

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

get_harvested_code_for_paper("1512.00567")
get_code_for_paper("1512.00567")
have("1512.00567")

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

curl https://syntology.ai/api/ran/1512.00567.json

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

[![Syntology run record](https://syntology.ai/api/ran/1512.00567.svg)](https://syntology.ai/paper/1512.00567)

Connect an agent — have() is free.