SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1807.10418 · 2018

W-TALC: Weakly-supervised Temporal Activity Localization and Classification

arXiv · PDF · Open in the Atlas

Code that ran

7 of 7 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 6 have no contract verdict; 0 failures 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/1807.10418.json · badge: /api/ran/1807.10418.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
sujoyp/wtalc-pytorch canonical (classifier label)
classifier basis: census
7 ran · 0 no recorded run
FunctionStatusWhere it lives
filter_segments Ran
ran on a synthesized input
sujoyp/wtalc-pytorch/detectionMAP.py
code served (permissive licence) · get_code("7758772e75188f8a")
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 (0,) 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 segment_predict: shape (2, 3) where columns are [video_idx, start, end]
segment_predict = np.array([
    [0, 10, 20],  # video 0, frames 10-20
    [1, 5, 15]    # video 1, frames 5-15
], dtype=float)

# Construct videonames: list mapping video indices to video names
videonames = ["video_A", "video_B"]

# Construct ambilist: list of tuples (videoname, _, start_time, end_time)
# The function multiplies start/end times by factor to get frame ranges
ambilist = [
    ("video_A", "dummy", 1.0, 2.5),  # video_A: frames 10-25 at factor=10
    ("video_B", "dummy", 0.5, 1.0)   # video_B: frames 5-10 at factor=10
]

# Construct factor: multiplier for converting time to frames
factor = 10

ARGS = {
    "segment_predict": segment_predict,
    "videonames": videonames,
    "ambilist": ambilist,
    "factor": factor
}
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 (0,) — shape (0,) 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
getAP Ran
ran on a synthesized input
sujoyp/wtalc-pytorch/classificationMAP.py
code served (permissive licence) · get_code("227942c6bc7e435b")
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

# Construct valid arguments for getAP function
# The function expects:
# - conf: array of confidence scores (float values between 0 and 1)
# - labels: array of binary labels (0 or 1)

# Create confidence scores and labels
np.random.seed(0)
conf = np.array([0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2], dtype='float32')
labels = np.array([1, 1, 0, 1, 0, 1, 0, 0], dtype='int32')

ARGS = {
    'conf': conf,
    'labels': labels
}
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 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
getClassificationMAP Ran
ran on a synthesized input
sujoyp/wtalc-pytorch/classificationMAP.py
code served (permissive licence) · get_code("b0f04cc9e7b36e90")
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

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

# Create small sample data
n_samples = 5
n_labels = 3

# confidence: n_samples x n_labels matrix of confidence scores (floats between 0 and 1)
confidence = np.random.rand(n_samples, n_labels).astype('float32')

# labels: n_samples x n_labels matrix of binary labels (0 or 1)
labels = np.random.randint(0, 2, size=(n_samples, n_labels)).astype('int32')

ARGS = {
    "confidence": confidence,
    "labels": labels
}
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 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
smooth Ran
ran on a synthesized input
sujoyp/wtalc-pytorch/detectionMAP.py
code served (permissive licence) · get_code("4a6f77bd7b90ba28")
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
import torch.nn as nn
import numpy as np

torch.manual_seed(0)

# The function `smooth(v)` takes a parameter `v` and returns it unchanged.
# Since the function body is just `return v`, it accepts any value.
# We'll provide a tensor as the most common use case.

v = torch.randn(2, 4)

ARGS = {
    "v": v
}
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 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
str2ind Ran
ran on a synthesized input
sujoyp/wtalc-pytorch/detectionMAP.py
code served (permissive licence) · get_code("165d3974e17f952d")
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

torch.manual_seed(0)

# The function str2ind takes a categoryname and classlist
# It searches for the categoryname in classlist and returns the index of the first match

# categoryname should be a string
categoryname = "cat"

# classlist should be a list of strings that can be searched
classlist = ["dog", "cat", "bird", "fish"]

ARGS = {
    "categoryname": categoryname,
    "classlist": classlist,
}
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
strlist2indlist Ran
ran, honoured the contract we drafted
sujoyp/wtalc-pytorch/utils.py
code served (permissive licence) · get_code("6b5e3d7c24f0a360")
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 int; 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

torch.manual_seed(0)

def str2ind(categoryname, classlist):
    return [i for i in range(len(classlist)) if categoryname == classlist[i].decode('utf-8')][0]

def strlist2indlist(strlist, classlist):
    return [str2ind(s, classlist) for s in strlist]

strlist = ['cat', 'dog', 'cat']
classlist = [b'cat', b'dog', b'bird', b'fish']

ARGS = {
    'strlist': strlist,
    'classlist': classlist
}
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 int — 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
strlist2multihot Ran
ran on a synthesized input
sujoyp/wtalc-pytorch/utils.py
code served (permissive licence) · get_code("2a0daffba5477477")
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 (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

# The function expects:
# - strlist: a list of strings to convert to multi-hot encoding
# - classlist: a list of byte strings to search through

# Create a small classlist with byte-encoded strings
classlist = [b'cat', b'dog', b'bird', b'fish']

# Create a strlist with some of those category names
strlist = ['cat', 'bird', 'cat']

ARGS = {
    'strlist': strlist,
    'classlist': classlist
}
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 (4,) — shape (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

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 6 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 6 later) and 1 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
filter_segments 1 other paper
1 later
136640190 (2022 (year only))
Dual-Evidential Learning for Weakly-supervised Temporal Action Localization
getAP 1 other paper
1 later
136640190 (2022 (year only))
Dual-Evidential Learning for Weakly-supervised Temporal Action Localization
getClassificationMAP 1 other paper
1 later
136640190 (2022 (year only))
Dual-Evidential Learning for Weakly-supervised Temporal Action Localization
smooth 1 other paper
1 later
136640190 (2022 (year only))
Dual-Evidential Learning for Weakly-supervised Temporal Action Localization
str2ind 1 other paper
1 later
136640190 (2022 (year only))
Dual-Evidential Learning for Weakly-supervised Temporal Action Localization
strlist2indlist 1 other paper
1 later
136640190 (2022 (year only))
Dual-Evidential Learning for Weakly-supervised Temporal Action Localization

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

Most activity localization methods in the literature suffer from the burden of frame-wise annotation requirement. Learning from weak labels may be a potential solution towards reducing such manual labeling effort. Recent years have witnessed a substantial influx of tagged videos on the Internet, which can serve as a rich source of weakly-supervised training data. Specifically, the correlations between videos with similar tags can be utilized to temporally localize the activities. Towards this goal, we present W-TALC, a Weakly-supervised Temporal Activity Localization and Classification framework using only video-level labels. The proposed network can be divided into two sub-networks, namely the Two-Stream based feature extractor network and a weakly-supervised module, which we learn by optimizing two complimentary loss functions. Qualitative and quantitative results on two challenging datasets - Thumos14 and ActivityNet1.2, demonstrate that the proposed method is able to detect activities at a fine granularity and achieve better performance than current state-of-the-art methods.

For agents

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

get_harvested_code_for_paper("1807.10418")
get_code_for_paper("1807.10418")
have("1807.10418")

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

curl https://syntology.ai/api/ran/1807.10418.json

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

[![Syntology run record](https://syntology.ai/api/ran/1807.10418.svg)](https://syntology.ai/paper/1807.10418)

Connect an agent — have() is free.