SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1906.07589 · 2019

Learning with Average Precision: Training Image Retrieval with a Listwise Loss

arXiv · PDF · Open in the Atlas

Code that ran

4 of 8 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 4 have no contract verdict; 0 failures of our instrument; 4 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/1906.07589.json · badge: /api/ran/1906.07589.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
almazan/deep-image-retrieval community (classifier label)
classifier basis not recorded on the link
4 ran · 4 no recorded run
FunctionStatusWhere it lives
find_and_list_classes Ran
ran on a synthesized input
almazan/deep-image-retrieval/dirtorch/datasets/generic_func.py
code served (permissive licence) · get_code("99c00be09d915e3f")
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 torch
import numpy as np

torch.manual_seed(0)

# Construct arguments for find_and_list_classes
# The function expects:
# 1. labels: list of image labels (can be str, int, etc.)
# 2. cls_idx: dict or None mapping class names to indices

# Simple case with string labels
labels = ['cat', 'dog', 'cat', 'bird', 'dog', 'cat']

# Optional cls_idx dict - can be None or a dict with some pre-assigned classes
cls_idx = None

ARGS = {
    'labels': labels,
    'cls_idx': cls_idx,
}
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 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
find_relevants Ran
ran on a synthesized input
almazan/deep-image-retrieval/dirtorch/datasets/generic_func.py
code served (permissive licence) · get_code("05cccbbaf896e094")
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 defaultdict. 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.

from collections import defaultdict

torch_seed = __import__('torch').manual_seed(0)

# The function takes 'labels' as input
# It iterates through len(labels) and accesses labels[i]
# It uses the labels as dictionary keys
# labels must be ordered (not a set), so it could be a list, tuple, or other sequence

# Create a simple list of class labels
labels = ['cat', 'dog', 'cat', 'bird', 'dog', 'cat']

ARGS = {
    'labels': labels
}
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 defaultdict — a defaultdict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a defaultdict 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
not_none Ran
ran on a synthesized input
almazan/deep-image-retrieval/dirtorch/datasets/generic.py
code served (permissive licence) · get_code("069961f5275162e3")
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

torch.manual_seed(0)

ARGS = {
    "label": "example_label"
}
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
sim_to_dist Ran
ran on a synthesized input
almazan/deep-image-retrieval/dirtorch/loss.py
code served (permissive licence) · get_code("aa71f3e917de3d1d")
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 (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 torch

torch.manual_seed(0)

# The function takes a 'scores' tensor and applies mathematical operations to it
# scores should be a tensor with values in a valid range for the formula
# The formula is: 1 - sqrt(2.001 - 2 * scores)
# For this to be valid, 2.001 - 2 * scores must be >= 0
# So scores <= 1.0005

scores = torch.tensor([[0.5, 0.3, 0.2, 0.1],
                       [0.8, 0.4, 0.6, 0.9]], dtype=torch.float32)

ARGS = {
    "scores": scores
}
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 (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
deploy Not yet run
unverified (harvested, no recorded run)
almazan/deep-image-retrieval/dirtorch/datasets/dataset.py
code served (permissive licence) · get_code("2ad33ba942a940d1")
How it ran

no invocation recorded.

deploy_and_split Not yet run
unverified (harvested, no recorded run)
almazan/deep-image-retrieval/dirtorch/datasets/dataset.py
code served (permissive licence) · get_code("52b59ca7fd9b5d7a")
How it ran

no invocation recorded.

l2_normalize Not yet run
unverified (harvested, no recorded run)
almazan/deep-image-retrieval/dirtorch/nets/rmac_resnet.py
code served (permissive licence) · get_code("426de3b16db2fb92")
How it ran

no invocation recorded.

split Not yet run
unverified (harvested, no recorded run)
almazan/deep-image-retrieval/dirtorch/datasets/dataset.py
code served (permissive licence) · get_code("0d31580390476d85")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 8 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.

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

Image retrieval can be formulated as a ranking problem where the goal is to order database images by decreasing similarity to the query. Recent deep models for image retrieval have outperformed traditional methods by leveraging ranking-tailored loss functions, but important theoretical and practical problems remain. First, rather than directly optimizing the global ranking, they minimize an upper-bound on the essential loss, which does not necessarily result in an optimal mean average precision (mAP). Second, these methods require significant engineering efforts to work well, e.g. special pre-training and hard-negative mining. In this paper we propose instead to directly optimize the global mAP by leveraging recent advances in listwise loss formulations. Using a histogram binning approximation, the AP can be differentiated and thus employed to end-to-end learning. Compared to existing losses, the proposed method considers thousands of images simultaneously at each iteration and eliminates the need for ad hoc tricks. It also establishes a new state of the art on many standard retrieval benchmarks. Models and evaluation scripts have been made available at https://europe.naverlabs.com/Deep-Image-Retrieval/

For agents

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

get_harvested_code_for_paper("1906.07589")
get_code_for_paper("1906.07589")
have("1906.07589")

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

curl https://syntology.ai/api/ran/1906.07589.json

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

[![Syntology run record](https://syntology.ai/api/ran/1906.07589.svg)](https://syntology.ai/paper/1906.07589)

Connect an agent — have() is free.