SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2107.08430 · 2021

YOLOX: Exceeding YOLO Series in 2021

arXiv · PDF · Open in the Atlas

Code that ran

16 of 23 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; 1 honoured the contract we drafted, 0 violated it and 7 have no contract verdict; 8 failures of our instrument; 7 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/2107.08430.json · badge: /api/ran/2107.08430.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
2023-MindSpore-1/ms-code-31 community (classifier label)
classifier basis not recorded on the link
6 ran · 0 no recorded run
StephenStorm/YOLOX community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
apolloauto/apollo-model-yolox community (classifier label)
classifier basis not recorded on the link
2 ran · 4 no recorded run
DataXujing/YOLOX- community (classifier label)
classifier basis not recorded on the link
2 ran · 1 no recorded run
kisna-aryan/YOLOX community (classifier label)
classifier basis not recorded on the link
1 ran · 2 no recorded run
chenyicai-0611/YOLOX-Flask-deployment community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
MegEngine/YOLOX community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
FunctionStatusWhere it lives
calculate_gain Ran
ran, honoured the contract we drafted
2023-MindSpore-1/ms-code-31/src/initializer.py
code served (permissive licence) · get_code("e460868d0e106ef7")
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 torch
import torch.nn as nn
import math

torch.manual_seed(0)

# Test case 1: linear nonlinearity
ARGS = {
    'nonlinearity': 'relu',
    'param': None
}
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 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
create_yolox_model Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
apolloauto/apollo-model-yolox/yolox/models/build.py
code served (permissive licence) · get_code("7111f135eb8bfff5")
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 Sequential. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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
from torch import nn

torch.manual_seed(0)

# Mock the yolox module structure since we can't import it
class MockExp:
    def __init__(self, name=None):
        self.num_classes = 80
        self.name = name
    
    def get_model(self):
        # Return a minimal nn.Module that has load_state_dict capability
        model = nn.Sequential(
            nn.Conv2d(3, 16, 3),
            nn.ReLU(),
            nn.Conv2d(16, 80, 1)
        )
        return model

def mock_get_exp(exp_name=None, exp_file=None):
    return MockExp(name=exp_name)

# Monkey-patch the yolox module
import sys
from types import ModuleType

yolox_module = ModuleType('yolox')
yolox_exp_module = ModuleType('yolox.exp')
yolox_exp_module.get_exp = mock_get_exp
yolox_exp_module.Exp = MockExp
yolox_module.exp = yolox_exp_module
sys.modules['yolox'] = yolox_module
sys.modules['yolox.exp'] = yolox_exp_module

ARGS = {
    'name': 'yolox-s',
    'pretrained': False,
    'num_classes': 80,
    'device': 'cpu',
    'exp_path': None,
    'ckpt_path': None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, yolox — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
create_yolox_model Ran
ran on a synthesized input
kisna-aryan/YOLOX/yolox/models/build.py
code served (permissive licence) · get_code("1c1d860f1cb790da")
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 MockYOLOXModel. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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
from torch import nn

torch.manual_seed(0)

# Mock the yolox module structure
class MockYOLOXModel(nn.Module):
    def __init__(self, num_classes=80):
        super().__init__()
        self.num_classes = num_classes
        self.linear = nn.Linear(10, num_classes)
    
    def forward(self, x):
        return self.linear(x)
    
    def load_state_dict(self, state_dict, strict=True):
        # Mock load_state_dict to accept any dictionary
        return super().load_state_dict(state_dict, strict=False)

class MockExp:
    def __init__(self, exp_name):
        self.exp_name = exp_name
        self.num_classes = 80
    
    def get_model(self):
        return MockYOLOXModel(num_classes=self.num_classes)

def mock_get_exp(exp_name):
    return MockExp(exp_name)

# Patch the yolox module
import sys
from unittest.mock import MagicMock

yolox_module = MagicMock()
yolox_module.exp.get_exp = mock_get_exp
yolox_module.exp.Exp = MockExp
sys.modules['yolox'] = yolox_module
sys.modules['yolox.exp'] = yolox_module.exp

ARGS = {
    'name': 'yolox-nano',
    'pretrained': False,
    'num_classes': 80,
    'device': 'cpu'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, yolox — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type MockYOLOXModel — a MockYOLOXModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a MockYOLOXModel 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
fuse_conv_and_bn Ran
ran on a synthesized input
DataXujing/YOLOX-/yolox/utils/model_utils.py
code served (permissive licence) · get_code("d3111a0ae5602114")
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 Conv2d. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a Conv2d layer
conv = nn.Conv2d(
    in_channels=3,
    out_channels=4,
    kernel_size=3,
    stride=1,
    padding=1,
    groups=1,
    bias=True
)

# Create a BatchNorm2d layer with same output channels as conv
bn = nn.BatchNorm2d(num_features=4)

# Initialize BatchNorm parameters to have valid values
bn.weight.data = torch.ones(4)
bn.bias.data = torch.zeros(4)
bn.running_mean.data = torch.zeros(4)
bn.running_var.data = torch.ones(4)
bn.eps = 1e-5

ARGS = {
    "conv": conv,
    "bn": bn
}
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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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
fuse_conv_and_bn Ran
ran on a synthesized input
apolloauto/apollo-model-yolox/yolox/utils/model_utils.py
code served (permissive licence) · get_code("57bfa676482bc295")
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 Conv2d. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a simple Conv2d layer
conv = nn.Conv2d(
    in_channels=3,
    out_channels=4,
    kernel_size=3,
    stride=1,
    padding=1,
    groups=1,
    bias=True
)

# Create a BatchNorm2d layer with the same output channels as conv
bn = nn.BatchNorm2d(num_features=4)

# Initialize batchnorm with some statistics so it has valid running_mean and running_var
# This is done by calling it on some dummy data
dummy_input = torch.randn(2, 3, 8, 8)
with torch.no_grad():
    _ = bn(conv(dummy_input))

ARGS = {
    "conv": conv,
    "bn": bn
}
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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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_activation Ran
ran; our contract draft was wrong, not the code
DataXujing/YOLOX-/yolox/models/network_blocks.py
code served (permissive licence) · get_code("6e3d5f4c3ce305d7")
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 SiLU. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'name': 'silu',
    'inplace': True,
}
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 SiLU — a SiLU output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SiLU 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_image_list Ran
ran on a synthesized input
chenyicai-0611/YOLOX-Flask-deployment/YOLOX-flask/backend/predict.py
code served (permissive licence) · get_code("2f7db76440582d33")
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 os
import tempfile
import shutil

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

# Create subdirectories
subdir1 = os.path.join(temp_dir, "subdir1")
subdir2 = os.path.join(temp_dir, "subdir2")
os.makedirs(subdir1, exist_ok=True)
os.makedirs(subdir2, exist_ok=True)

# Create dummy image files with various extensions
open(os.path.join(temp_dir, "image1.jpg"), 'w').close()
open(os.path.join(temp_dir, "image2.png"), 'w').close()
open(os.path.join(subdir1, "image3.jpeg"), 'w').close()
open(os.path.join(subdir1, "image4.webp"), 'w').close()
open(os.path.join(subdir2, "image5.bmp"), 'w').close()
open(os.path.join(temp_dir, "notimage.txt"), 'w').close()

ARGS = {
    'path': 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 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
get_logger Ran
ran on a synthesized input
2023-MindSpore-1/ms-code-31/src/logger.py
code served (permissive licence) · get_code("bb879be65ea0ad0d")
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 LOGGER. 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 os
import sys
import logging
import tempfile
from datetime import datetime

# Create a temporary directory for logging
temp_log_dir = tempfile.mkdtemp()

ARGS = {
    'path': temp_log_dir,
    'rank': 0
}
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 LOGGER — a LOGGER output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a LOGGER 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
kaiming_uniform_ Ran
ran; our contract draft was wrong, not the code
2023-MindSpore-1/ms-code-31/src/initializer.py
code served (permissive licence) · get_code("ac69eca3db22245e")
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 (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
import math

# Create a small numpy array with shape (4, 3) to represent a weight matrix
# The function expects at least 2 dimensions
arr = np.empty((4, 3))

# Define arguments for kaiming_uniform_
ARGS = {
    'arr': arr,
    'a': 0.2,
    'mode': 'fan_in',
    'nonlinearity': 'leaky_relu'
}
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 (4, 3) — shape (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
2023-MindSpore-1/ms-code-31/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
2023-MindSpore-1/ms-code-31/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
2023-MindSpore-1/ms-code-31/model_utils/config.py
code served (permissive licence) · get_code("a5a844a8ffaaa722")
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 configuration
temp_dir = tempfile.mkdtemp()
yaml_path = os.path.join(temp_dir, "config.yaml")

# Write a valid multi-document YAML file
yaml_content = """
model:
  name: test_model
  layers: 3
training:
  epochs: 10
  batch_size: 32
---
model: "Model configuration"
training: "Training hyperparameters"
---
model:
  name: ["model_a", "model_b"]
  layers: [2, 3, 4]
"""

with open(yaml_path, 'w', encoding='utf-8') 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
reduce_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
StephenStorm/YOLOX/yolox/models/VariFocalLoss.py
code served (permissive licence) · get_code("93358ba7aa223343")
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 Tensor, 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
import torch.nn.functional as F

torch.manual_seed(0)

# Create a small elementwise loss tensor
loss = torch.randn(2, 4)

# Test with different reduction modes
ARGS = {
    "loss": loss,
    "reduction": "mean"
}
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 Tensor, 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
replace_module Ran
ran on a synthesized input
MegEngine/YOLOX/yolox/utils/model_utils.py
code served (permissive licence) · get_code("14dbd04931eace27")
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 SimpleModule. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a simple module with some children to test replacement
class SimpleModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.relu = nn.ReLU()
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.relu(x)
        x = self.linear2(x)
        return x

class ReplacementModule(nn.Module):
    def __init__(self):
        super().__init__()
    
    def forward(self, x):
        return x

# Create the main module with ReLU layers to replace
module = SimpleModule()

# Define the type to replace
replaced_module_type = nn.ReLU

# Define the new type
new_module_type = nn.Identity

# Define a custom replace function
def custom_replace_func(replaced_module_type, new_module_type):
    return new_module_type()

ARGS = {
    "module": module,
    "replaced_module_type": replaced_module_type,
    "new_module_type": new_module_type,
    "replace_func": custom_replace_func
}
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 SimpleModule — a SimpleModule output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleModule 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
varifocal_loss Ran
ran; our contract draft was wrong, not the code
StephenStorm/YOLOX/yolox/models/VariFocalLoss.py
code served (permissive licence) · get_code("48e94cb9aba6d73d")
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 Tensor, 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

torch.manual_seed(0)

# Create tiny tensors for the function
batch_size = 2
num_classes = 4

# pred: prediction logits with shape (N, C)
pred = torch.randn(batch_size, num_classes)

# target: learning target (iou-aware classification scores) with shape (N, C)
target = torch.rand(batch_size, num_classes)

# weight: optional element-wise weights, same shape as loss
weight = torch.ones(batch_size, num_classes)

# Scalar parameters
alpha = 0.75
gamma = 2.0
iou_weighted = True
reduction = 'mean'
avg_factor = None

ARGS = {
    'pred': pred,
    'target': target,
    'weight': weight,
    'alpha': alpha,
    'gamma': gamma,
    'iou_weighted': iou_weighted,
    'reduction': reduction,
    'avg_factor': avg_factor,
}
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 Tensor, 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
weight_reduce_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
StephenStorm/YOLOX/yolox/models/VariFocalLoss.py
code served (permissive licence) · get_code("8553862254d49840")
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 Tensor, 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
import torch.nn.functional as F

torch.manual_seed(0)

# Create tiny tensors for the function parameters
loss_tensor = torch.randn(2, 4)  # Elementwise loss tensor
weight_tensor = torch.ones(2, 4)  # Element-wise weights

ARGS = {
    'loss': loss_tensor,
    'weight': weight_tensor,
    'reduction': 'mean',
    'avg_factor': None
}
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 Tensor, 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
fuse_model Not yet run
unverified (harvested, no recorded run)
DataXujing/YOLOX-/yolox/utils/model_utils.py
code served (permissive licence) · get_code("bce7fbf2f1717430")
How it ran

no invocation recorded.

fuse_model Not yet run
unverified (harvested, no recorded run)
apolloauto/apollo-model-yolox/yolox/utils/model_utils.py
code served (permissive licence) · get_code("927238a3415f3aa2")
How it ran

no invocation recorded.

get_model_info Not yet run
unverified (harvested, no recorded run)
apolloauto/apollo-model-yolox/yolox/utils/model_utils.py
code served (permissive licence) · get_code("624c83e485703cd7")
How it ran

no invocation recorded.

yolox_nano Not yet run
unverified (harvested, no recorded run)
apolloauto/apollo-model-yolox/yolox/models/build.py
code served (permissive licence) · get_code("706978d51729ca8d")
How it ran

no invocation recorded.

yolox_nano Not yet run
unverified (harvested, no recorded run)
kisna-aryan/YOLOX/yolox/models/build.py
code served (permissive licence) · get_code("4657da89fcf077b4")
How it ran

no invocation recorded.

yolox_tiny Not yet run
unverified (harvested, no recorded run)
apolloauto/apollo-model-yolox/yolox/models/build.py
code served (permissive licence) · get_code("af7e14a4cec878b8")
How it ran

no invocation recorded.

yolox_tiny Not yet run
unverified (harvested, no recorded run)
kisna-aryan/YOLOX/yolox/models/build.py
code served (permissive licence) · get_code("6bbd9ef4e16fa647")
How it ran

no invocation recorded.

Also present in other papers' code

Of 23 samples lifted from this paper's linked repositories, 23 carry a function identity and 0 do not. Of the 23 distinct functions with an identity, 13 are also present in other papers whose code we hold (9 with at least one paper dated earlier, 0 the same month, 9 later) and 10 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
24 dated earlier, 4 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
1512.00567 (2015-12)
Rethinking the Inception Architecture for Computer Vision
and 25 more not named here
parse_cli_to_yaml 20 other papers
17 dated earlier, 3 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
1512.00567 (2015-12)
Rethinking the Inception Architecture for Computer Vision
and 17 more not named here
reduce_loss 18 other papers
2 dated earlier, 16 later
2003.10152 (2020-03)
SOLOv2: Dynamic and Fast Instance Segmentation
2003.10778 (2020-03)
PanNuke Dataset Extension, Insights and Baselines
2108.06747 (2021-08)
SOTR: Segmenting Objects with Transformers
and 15 more not named here
get_activation 10 other papers
1 dated earlier, 9 later
1803.11365 (2018-03)
Cross-Domain Weakly-Supervised Object Detection through Progressive Domain Adaptation
2108.04230 (2021-08)
Workshop on Autonomous Driving at CVPR 2021: Technical Report for Streaming Perception Challenge
2110.06864 (2021-10)
ByteTrack: Multi-Object Tracking by Associating Every Detection Box
and 7 more not named here
weight_reduce_loss 7 other papers
2 dated earlier, 5 later
2003.10152 (2020-03)
SOLOv2: Dynamic and Fast Instance Segmentation
2003.10778 (2020-03)
PanNuke Dataset Extension, Insights and Baselines
2023.findings-emnlp.517 (2023 (year only))
Text2Tree: Aligning Text Representation to the Label Tree Hierarchy for Imbalanced Medical Classification
and 4 more not named here
calculate_gain 3 other papers
3 dated earlier
1804.02767 (2018-04)
YOLOv3: An Incremental Improvement
2104.00298 (2021-04)
EfficientNetV2: Smaller Models and Faster Training
2105.15203 (2021-05)
SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers
create_yolox_model 2 other papers
2 later
2108.04230 (2021-08)
Workshop on Autonomous Driving at CVPR 2021: Technical Report for Streaming Perception Challenge
2601.08797 (2026-01)
Dentalx: Context-Aware Dental Disease Detection With Radiographs
kaiming_uniform_ 2 other papers
2 dated earlier
1804.02767 (2018-04)
YOLOv3: An Incremental Improvement
2104.00298 (2021-04)
EfficientNetV2: Smaller Models and Faster Training
varifocal_loss 2 other papers
2 later
Deng_Harmonious_Teacher_for_Cross-Domain_Object_Detection_CVPR_2023_paper (2023 (year only))
Harmonious Teacher for Cross-domain Object Detection
2403.09359 (2024-03)
D3T: Distinctive Dual-Domain Teacher Zigzagging Across RGB-Thermal Gap for Domain-Adaptive Object Detection
yolox_nano 2 other papers
2 later
2108.04230 (2021-08)
Workshop on Autonomous Driving at CVPR 2021: Technical Report for Streaming Perception Challenge
2601.08797 (2026-01)
Dentalx: Context-Aware Dental Disease Detection With Radiographs
yolox_tiny 2 other papers
2 later
2108.04230 (2021-08)
Workshop on Autonomous Driving at CVPR 2021: Technical Report for Streaming Perception Challenge
2601.08797 (2026-01)
Dentalx: Context-Aware Dental Disease Detection With Radiographs
fuse_conv_and_bn 1 other paper
1 dated earlier
1803.11365 (2018-03)
Cross-Domain Weakly-Supervised Object Detection through Progressive Domain Adaptation
fuse_model 1 other paper
1 dated earlier
1803.11365 (2018-03)
Cross-Domain Weakly-Supervised Object Detection through Progressive Domain Adaptation

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

In this report, we present some experienced improvements to YOLO series, forming a new high-performance detector -- YOLOX. We switch the YOLO detector to an anchor-free manner and conduct other advanced detection techniques, i.e., a decoupled head and the leading label assignment strategy SimOTA to achieve state-of-the-art results across a large scale range of models: For YOLO-Nano with only 0.91M parameters and 1.08G FLOPs, we get 25.3% AP on COCO, surpassing NanoDet by 1.8% AP; for YOLOv3, one of the most widely used detectors in industry, we boost it to 47.3% AP on COCO, outperforming the current best practice by 3.0% AP; for YOLOX-L with roughly the same amount of parameters as YOLOv4-CSP, YOLOv5-L, we achieve 50.0% AP on COCO at a speed of 68.9 FPS on Tesla V100, exceeding YOLOv5-L by 1.8% AP. Further, we won the 1st Place on Streaming Perception Challenge (Workshop on Autonomous Driving at CVPR 2021) using a single YOLOX-L model. We hope this report can provide useful experience for developers and researchers in practical scenes, and we also provide deploy versions with ONNX, TensorRT, NCNN, and Openvino supported. Source code is at https://github.com/Megvii-BaseDetection/YOLOX.

For agents

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

get_harvested_code_for_paper("2107.08430")
get_code_for_paper("2107.08430")
have("2107.08430")

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

curl https://syntology.ai/api/ran/2107.08430.json

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

[![Syntology run record](https://syntology.ai/api/ran/2107.08430.svg)](https://syntology.ai/paper/2107.08430)

Connect an agent — have() is free.