SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1910.09862 · 2019

A Prototypical Triplet Loss for Cover Detection

arXiv · PDF · Open in the Atlas

Code that ran

3 of 6 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 2 have no contract verdict; 1 failure of our instrument; 3 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/1910.09862.json · badge: /api/ran/1910.09862.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
gdoras/PrototypicalTripletLoss canonical (classifier label)
classifier basis: census
3 ran · 3 no recorded run
FunctionStatusWhere it lives
masked_maximum Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
gdoras/PrototypicalTripletLoss/losses/prototypical_triplet_loss.py
code served (permissive licence) · get_code("a996b3938f8e4e9f")
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, 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 tensorflow as tf
import numpy as np

# Set seeds for reproducibility
tf.random.set_seed(0)
np.random.seed(0)

# Create small test tensors
batch_size = 2
feature_dim = 4

# Create a 2-D float tensor of size [n, m]
data = tf.constant([
    [1.0, 2.0, 3.0, 4.0],
    [5.0, 1.0, 2.0, 3.0]
], dtype=tf.float32)

# Create a 2-D boolean mask of size [n, m]
mask = tf.constant([
    [True, False, True, False],
    [True, True, False, True]
], dtype=tf.bool)

# Convert mask to float for the multiply operation
mask = tf.cast(mask, tf.float32)

ARGS = {
    "data": data,
    "mask": mask,
    "dim": 1
}
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, 1) — shape (2, 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
masked_minimum Ran
ran on a synthesized input
gdoras/PrototypicalTripletLoss/losses/prototypical_triplet_loss_tf2.py
code served (permissive licence) · get_code("11ebe8921019da89")
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, 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 tensorflow as tf
import numpy as np

tf.random.set_seed(0)

# Create small 2-D tensors for testing
# data: 2-D float Tensor of size [n, m]
data = tf.constant([
    [1.0, 2.0, 3.0, 4.0],
    [5.0, 6.0, 7.0, 8.0]
], dtype=tf.float32)

# mask: 2-D Boolean Tensor of size [n, m]
mask = tf.constant([
    [True, False, True, False],
    [False, True, False, True]
], dtype=tf.bool)

# Convert mask to float for multiplication
mask_float = tf.cast(mask, tf.float32)

ARGS = {
    "data": data,
    "mask": mask_float,
    "dim": 1
}
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, 1) — shape (2, 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
prototypical_triplet_loss Ran
ran on a synthesized input
gdoras/PrototypicalTripletLoss/losses/prototypical_triplet_loss_tf2.py
code served (permissive licence) · get_code("e3f08a299e6b1fbf")
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 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 tensorflow as tf
import numpy as np

# Set seeds for reproducibility
tf.random.set_seed(0)
np.random.seed(0)

# Create small test data
batch_size = 4
embedding_dim = 8

# Create embeddings: small batch with normalized vectors
embeddings_np = np.random.randn(batch_size, embedding_dim).astype(np.float32)
# L2 normalize embeddings
embeddings_np = embeddings_np / (np.linalg.norm(embeddings_np, axis=1, keepdims=True) + 1e-8)
embeddings = tf.constant(embeddings_np)

# Create labels: some samples from same cluster, some different
# Labels: [0, 0, 1, 1] - two clusters with 2 samples each
labels = tf.constant([0, 0, 1, 1], dtype=tf.int32)

# Define the arguments dictionary
ARGS = {
    "y_true": labels,
    "y_pred": embeddings,
    "margin": 1.0
}
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 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
masked_maximum Not yet run
unverified (harvested, no recorded run)
gdoras/PrototypicalTripletLoss/losses/prototypical_triplet_loss_tf2.py
code served (permissive licence) · get_code("20a17a513f7a3e4d")
How it ran

no invocation recorded.

masked_minimum Not yet run
unverified (harvested, no recorded run)
gdoras/PrototypicalTripletLoss/losses/prototypical_triplet_loss.py
code served (permissive licence) · get_code("26ad6d209dc9fe92")
How it ran

no invocation recorded.

prototypical_triplet_loss Not yet run
unverified (harvested, no recorded run)
gdoras/PrototypicalTripletLoss/losses/prototypical_triplet_loss.py
code served (permissive licence) · get_code("4ff7c274d8408f82")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 4 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 3 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
masked_maximum 1 other paper
1 dated earlier
1907.10164 (2019-07)
Cap2Det: Learning to Amplify Weak Caption Supervision for Object Detection

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

Automatic cover detection -- the task of finding in a audio dataset all covers of a query track -- has long been a challenging theoretical problem in MIR community. It also became a practical need for music composers societies requiring to detect automatically if an audio excerpt embeds musical content belonging to their catalog. In a recent work, we addressed this problem with a convolutional neural network mapping each track's dominant melody to an embedding vector, and trained to minimize cover pairs distance in the embeddings space, while maximizing it for non-covers. We showed in particular that training this model with enough works having five or more covers yields state-of-the-art results. This however does not reflect the realistic use case, where music catalogs typically contain works with zero or at most one or two covers. We thus introduce here a new test set incorporating these constraints, and propose two contributions to improve our model's accuracy under these stricter conditions: we replace dominant melody with multi-pitch representation as input data, and describe a novel prototypical triplet loss designed to improve covers clustering. We show that these changes improve results significantly for two concrete use cases, large dataset lookup and live songs identification.

For agents

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

get_harvested_code_for_paper("1910.09862")
get_code_for_paper("1910.09862")
have("1910.09862")

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

curl https://syntology.ai/api/ran/1910.09862.json

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

[![Syntology run record](https://syntology.ai/api/ran/1910.09862.svg)](https://syntology.ai/paper/1910.09862)

Connect an agent — have() is free.