SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2011.01447 · 2020

A Two-Stage Approach to Device-Robust Acoustic Scene Classification

arXiv · PDF · Open in the Atlas

Code that ran

10 of 10 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 9 have no contract verdict; 1 failure of our instrument; 0 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/2011.01447.json · badge: /api/ran/2011.01447.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
MihawkHu/DCASE2020_task1 canonical (classifier label)
classifier basis: census
10 ran · 0 no recorded run
FunctionStatusWhere it lives
My_freq_split1 Ran
ran on a synthesized input
MihawkHu/DCASE2020_task1/other_TF2_format_pretrained/resnet.py
code served (permissive licence) · get_code("b7b59b8162d03a3f")
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 4, dtype family float, shape (2, 64, 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 torch

torch.manual_seed(0)

# The function My_freq_split1 takes a tensor x and slices it
# to get elements from index 0:64 along dimension 1
# It expects a 4D tensor (batch, channels/freq, height, width)

x = torch.randn(2, 128, 4, 3)  # batch=2, freq_dim=128, height=4, width=3

ARGS = {
    "x": x
}
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 4, dtype family float, shape (2, 64, 4, 3) — shape (2, 64, 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
channel_attention Ran
ran; our contract draft was wrong, not the code
MihawkHu/DCASE2020_task1/other_TF2_format_pretrained/attention_layer.py
code served (permissive licence) · get_code("6d7713d3d9944832")
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 KerasTensor. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
from tensorflow.keras.layers import Input
import numpy as np

# Set up a small input tensor for the channel_attention function
# The function expects a 4D tensor (batch, height, width, channels) in channels_last format
# We'll use batch_size=2, height=4, width=3, channels=8

batch_size = 2
height = 4
width = 3
channels = 8

# Create a small random input tensor
input_data = tf.random.normal((batch_size, height, width, channels))

# Create the input layer (required for Keras functional API)
input_feature = Input(shape=(height, width, channels))

ARGS = {
    'input_feature': input_feature,
    'ratio': 8
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type KerasTensor — a KerasTensor 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
cmvn Ran
ran on a synthesized input
MihawkHu/DCASE2020_task1/task1b/funcs.py
code served (permissive licence) · get_code("dc9e7eb5bd140a8c")
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, 3, 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 numpy as np

# Create sample data for the cmvn function
# The function expects data with shape (batch_size, height, width, 1) or similar
# It processes each batch item by squeezing, transposing, normalizing, and reshaping

# Create a small batch of data: shape (2, 4, 3, 1)
# batch=2, height=4, width=3, channels=1
data = np.random.randn(2, 4, 3, 1).astype(np.float32)

ARGS = {
    "data": data
}
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, 3, 1) — shape (2, 4, 3, 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
conv_layer1 Ran
ran on a synthesized input
MihawkHu/DCASE2020_task1/other_TF2_format_pretrained/fcnn_att.py
code served (permissive licence) · get_code("e8e40a5d7c8a914b")
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 KerasTensor. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

# Create a small input tensor with shape (batch_size, height, width, channels)
# batch=2, height=4, width=3, channels=1
inputs = tf.keras.Input(shape=(4, 3, 1))

ARGS = {
    'inputs': inputs,
    'num_channels': 6,
    'num_filters': 14,
    'learn_bn': True,
    'wd': 0.0001,
    'use_relu': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type KerasTensor — a KerasTensor 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
freq_split1 Ran
ran on a synthesized input
MihawkHu/DCASE2020_task1/other_TF2_format_pretrained/mobnet.py
code served (permissive licence) · get_code("7a928b2854b78f59")
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 4, dtype family float, shape (2, 64, 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 torch

torch.manual_seed(0)

# freq_split1 expects a 4D tensor and slices it along dimension 1 (indices 0:64)
# Shape: (batch, channels, height, width)
x = torch.randn(2, 128, 4, 3)

ARGS = {
    "x": x
}
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 4, dtype family float, shape (2, 64, 4, 3) — shape (2, 64, 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
frequency_masking Ran
ran on a synthesized input
MihawkHu/DCASE2020_task1/task1b/funcs.py
code served (permissive licence) · get_code("e91b10240378b5b3")
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 2, dtype family float, shape (16, 8) 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 random

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

# Create a small mel-spectrogram tensor
# Shape: (frequency_bins, time_steps)
mel_spectrogram = torch.randn(16, 8)

ARGS = {
    "mel_spectrogram": mel_spectrogram,
    "frequency_masking_para": 13,
    "frequency_mask_num": 1
}
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 2, dtype family float, shape (16, 8) — shape (16, 8) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
pad_depth Ran
ran on a synthesized input
MihawkHu/DCASE2020_task1/other_TF2_format_pretrained/mobnet.py
code served (permissive licence) · get_code("9938cdab9fb51473")
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 EagerTensor, ndim 4, dtype family float, shape (2, 4, 3, 5) 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 tensorflow as tf
import numpy as np

# Create a small test tensor using TensorFlow
# The function uses K.zeros_like which requires a tensor input
inputs = tf.constant(np.random.randn(2, 4, 3, 5).astype(np.float32))

# desired_channels is an integer representing the number of channels to pad to
desired_channels = 8

ARGS = {
    'inputs': inputs,
    'desired_channels': desired_channels
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 4, dtype family float, shape (2, 4, 3, 5) — shape (2, 4, 3, 5) 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
relu6 Ran
ran on a synthesized input
MihawkHu/DCASE2020_task1/other_TF2_format_pretrained/fcnn_att.py
code served (permissive licence) · get_code("12cf81ec779a163f")
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 EagerTensor, ndim 2, dtype family float, shape (2, 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 tensorflow as tf
from tensorflow.keras import backend as K
import numpy as np

# Create a small test tensor using TensorFlow
x = tf.constant([[1.0, -2.0, 3.0, -4.0],
                  [5.0, -6.0, 7.0, -8.0]], dtype=tf.float32)

ARGS = {
    'x': x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 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
resnet_layer Ran
ran on a synthesized input
MihawkHu/DCASE2020_task1/other_TF2_format_pretrained/mobnet.py
code served (permissive licence) · get_code("c27af96c652b89af")
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 KerasTensor. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

# Create a small input tensor (batch_size=2, height=4, width=4, channels=3)
inputs = tf.keras.Input(shape=(4, 4, 3))

ARGS = {
    'inputs': inputs,
    'num_filters': 16,
    'kernel_size': 3,
    'strides': 1,
    'learn_bn': True,
    'wd': 0.0001,
    'use_relu': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type KerasTensor — a KerasTensor 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
time_masking Ran
ran on a synthesized input
MihawkHu/DCASE2020_task1/task1b/funcs.py
code served (permissive licence) · get_code("7e6103127d4b6834")
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 2, dtype family float, shape (4, 50) 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 random

torch.manual_seed(0)

# Create a small mel_spectrogram tensor
# Shape: (mel_bins, time_steps) - using distinct dimensions for clarity
mel_spectrogram = torch.randn(4, 50)

# Define ARGS with all parameters
ARGS = {
    'mel_spectrogram': mel_spectrogram,
    'time_masking_para': 40,
    'time_mask_num': 1
}
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 2, dtype family float, shape (4, 50) — shape (4, 50) 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

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 0 later) and 9 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
channel_attention 1 other paper
1 dated earlier
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

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

Abstract

To improve device robustness, a highly desirable key feature of a competitive data-driven acoustic scene classification (ASC) system, a novel two-stage system based on fully convolutional neural networks (CNNs) is proposed. Our two-stage system leverages on an ad-hoc score combination based on two CNN classifiers: (i) the first CNN classifies acoustic inputs into one of three broad classes, and (ii) the second CNN classifies the same inputs into one of ten finer-grained classes. Three different CNN architectures are explored to implement the two-stage classifiers, and a frequency sub-sampling scheme is investigated. Moreover, novel data augmentation schemes for ASC are also investigated. Evaluated on DCASE 2020 Task 1a, our results show that the proposed ASC system attains a state-of-the-art accuracy on the development set, where our best system, a two-stage fusion of CNN ensembles, delivers a 81.9% average accuracy among multi-device test data, and it obtains a significant improvement on unseen devices. Finally, neural saliency analysis with class activation mapping (CAM) gives new insights on the patterns learnt by our models.

For agents

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

get_harvested_code_for_paper("2011.01447")
get_code_for_paper("2011.01447")
have("2011.01447")

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

curl https://syntology.ai/api/ran/2011.01447.json

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

[![Syntology run record](https://syntology.ai/api/ran/2011.01447.svg)](https://syntology.ai/paper/2011.01447)

Connect an agent — have() is free.