SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.08951 · 2023

The Tiny Time-series Transformer: Low-latency High-throughput Classification of Astronomical Transients using Deep Model Compression

arXiv · PDF · Open in the Atlas

Code that ran

9 of 9 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; 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/2303.08951.json · badge: /api/ran/2303.08951.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
tallamjr/astronet canonical (classifier label)
classifier basis not recorded on the link
9 ran · 0 no recorded run
FunctionStatusWhere it lives
astronet_logger Ran
ran on a synthesized input
tallamjr/astronet/astronet/utils.py
code served (permissive licence) · get_code("842179f8245b9a72")
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 logging
import colorama

class CustomFormatter(logging.Formatter):
    grey = '\x1b[38;20m'
    yellow = '\x1b[33;20m'
    red = '\x1b[31;20m'
    bold_red = '\x1b[31;1m'
    reset = '\x1b[0m'
    white = colorama.Fore.WHITE
    FORMAT = '[%(asctime)s] '
    FORMAT += '{%(filename)s:%(lineno)d} '
    FORMAT += '%(levelname)s '
    FORMAT += '- %(message)s'
    FORMATS = {logging.DEBUG: grey + FORMAT + reset, logging.INFO: white + FORMAT + reset, logging.WARNING: yellow + FORMAT + reset, logging.ERROR: red + FORMAT + reset, logging.CRITICAL: bold_red + FORMAT + reset}

    def format(self, record):
        DATEFORMAT = '%y-%m-%d %H:%M:%S'
        log_fmt = self.FORMATS.get(record.levelno)
        formatter = logging.Formatter(log_fmt, datefmt=DATEFORMAT)
        return formatter.format(record)

def astronet_logger(name: str, level: str='INFO') -> logging.Logger:
    """Initialise python logger.

    Parameters
    ----------
    name : str
        Should be __name__ of file or module.
    log_level : str
        levels: DEBUG, INFO, WARNING, ERROR, CRITICAL, OFF

    Returns
    ----------
    logger : logging.Logger
        Python Logger

    Examples
    ----------
    >>> logger.debug('debug message')
    >>> logger.info('info message')
    >>> logger.warning('warn message')
    >>> logger.error('error message')
    >>> logger.critical('critical message')
    """
    logger = logging.getLogger(name)
    logger.setLevel(level)
    ch = logging.StreamHandler()
    ch.setLevel(level)
    ch.setFormatter(CustomFormatter())
    logger.addHandler(ch)
    logger.propagate = False
    return logger

ARGS = {
    'name': 'test_module',
    'level': 'INFO'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
colorama — 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
check_size Ran
ran on a synthesized input
tallamjr/astronet/astronet/dutils.py
code served (permissive licence) · get_code("59a8eb6b9570d35a")
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 str. 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 subprocess
import tempfile
import os

# Create a temporary directory and file for testing
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test_file.txt")

# Write some content to the file so it has a measurable size
with open(temp_file, 'w') as f:
    f.write("test content " * 100)

ARGS = {
    "filepath": temp_file
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
custom_log_loss Ran
ran on a synthesized input
tallamjr/astronet/astronet/metrics.py
code served (permissive licence) · get_code("534e0543c996d4a0")
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 numpy as np
import tensorflow as tf

# Create y_true as a TensorFlow tensor (batch_size=2, num_classes=4)
# Multi-hot encoded labels
y_true = tf.constant([
    [1.0, 0.0, 1.0, 0.0],
    [0.0, 1.0, 1.0, 0.0]
], dtype=tf.float32)

# Create y_pred as a numpy array with predicted probabilities (batch_size=2, num_classes=4)
y_pred = np.array([
    [0.7, 0.1, 0.15, 0.05],
    [0.1, 0.6, 0.25, 0.05]
], dtype=np.float32)

ARGS = {
    "y_true": y_true,
    "y_pred": y_pred
}
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
find_optimal_batch_size Ran
ran on a synthesized input
tallamjr/astronet/astronet/utils.py
code served (permissive licence) · get_code("e7c9f83f79ac50f7")
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 takes only one parameter: training_set_length (int)
# It determines optimal batch size based on the dataset length
# No complex objects needed

ARGS = {
    "training_set_length": 5000
}
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
lazy_load_plasticc_noZ Ran
ran on a synthesized input
tallamjr/astronet/astronet/datasets.py
code served (permissive licence) · get_code("04641fffa2cd84aa")
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 _FlatMapDataset. 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 small sample data
# X should be a list/iterable of samples (each can be a tensor or array)
# y should be a list/iterable of labels (each can be a tensor or array)

X = [
    tf.constant([[1.0, 2.0, 3.0, 4.0]], dtype=tf.float32),
    tf.constant([[5.0, 6.0, 7.0, 8.0]], dtype=tf.float32),
]

y = [
    tf.constant([0], dtype=tf.int32),
    tf.constant([1], dtype=tf.int32),
]

ARGS = {
    "X": X,
    "y": y,
}
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 _FlatMapDataset — a _FlatMapDataset output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a _FlatMapDataset 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
lazy_load_plasticc_wZ Ran
ran on a synthesized input
tallamjr/astronet/astronet/datasets.py
code served (permissive licence) · get_code("7297496d01f82deb")
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 _FlatMapDataset. 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 small sample data for the function
# X: list of input tensors (input_1)
# Z: list of auxiliary input tensors (input_2)
# y: list of labels

# Create tiny sample tensors
X = [
    tf.constant(np.random.randn(4, 3).astype(np.float32)),
    tf.constant(np.random.randn(4, 3).astype(np.float32))
]

Z = [
    tf.constant(np.random.randn(2, 5).astype(np.float32)),
    tf.constant(np.random.randn(2, 5).astype(np.float32))
]

y = [
    tf.constant(1.0, dtype=tf.float32),
    tf.constant(0.0, dtype=tf.float32)
]

ARGS = {
    'X': X,
    'Z': Z,
    'y': y
}
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 _FlatMapDataset — a _FlatMapDataset output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a _FlatMapDataset 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
lazy_load_tfdataset_from_numpy Ran
ran on a synthesized input
tallamjr/astronet/astronet/datasets.py
code served (permissive licence) · get_code("2b105e2b7e0c3b2c")
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 _FlatMapDataset. 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 numpy as np
import tempfile
import os

# Create a temporary numpy file with sample data
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "sample_data.npy")

# Create sample numpy array with shape (2, 4) - batch size 2, dimension 4
sample_data = np.random.randn(2, 4).astype(np.float32)
np.save(temp_file, sample_data)

ARGS = {
    "file": temp_file
}
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 _FlatMapDataset — a _FlatMapDataset output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a _FlatMapDataset 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
train_val_test_split Ran
ran on a synthesized input
tallamjr/astronet/astronet/utils.py
code served (permissive licence) · get_code("99fc3ce23c4cb98b")
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 pandas as pd
import numpy as np

# Create a small dataframe with sample data
np.random.seed(0)
data = {
    'feature_1': np.random.randn(100),
    'feature_2': np.random.randn(100),
    'feature_3': np.random.randn(100),
    'feature_4': np.random.randn(100),
    'extra_col': np.random.randn(100)
}
df = pd.DataFrame(data)

# Define the columns to use in the split
cols = ['feature_1', 'feature_2', 'feature_3', 'feature_4']

ARGS = {
    'df': df,
    'cols': cols
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — 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
zippify Ran
ran on a synthesized input
tallamjr/astronet/astronet/dutils.py
code served (permissive licence) · get_code("e169a3cd7fbfaf1a")
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 str. 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 tempfile

# Create a temporary directory and file for testing
temp_dir = tempfile.mkdtemp()
test_file_path = os.path.join(temp_dir, "test_file.txt")

# Write some content to the test file
with open(test_file_path, 'w') as f:
    f.write("This is test content for zipping.")

ARGS = {
    'file_path': test_file_path,
    'zipped_name': os.path.join(temp_dir, "output.zip")
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 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 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.

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

A new golden age in astronomy is upon us, dominated by data. Large astronomical surveys are broadcasting unprecedented rates of information, demanding machine learning as a critical component in modern scientific pipelines to handle the deluge of data. The upcoming Legacy Survey of Space and Time (LSST) of the Vera C. Rubin Observatory will raise the big-data bar for time-domain astronomy, with an expected 10 million alerts per-night, and generating many petabytes of data over the lifetime of the survey. Fast and efficient classification algorithms that can operate in real-time, yet robustly and accurately, are needed for time-critical events where additional resources can be sought for follow-up analyses. In order to handle such data, state-of-the-art deep learning architectures coupled with tools that leverage modern hardware accelerators are essential. We showcase how the use of modern deep compression methods can achieve a $18\times$ reduction in model size, whilst preserving classification performance. We also show that in addition to the deep compression techniques, careful choice of file formats can improve inference latency, and thereby throughput of alerts, on the order of $8\times$ for local processing, and $5\times$ in a live production setting. To test this in a live setting, we deploy this optimised version of the original time-series transformer, t2, into the community alert broking system of FINK on real Zwicky Transient Facility (ZTF) alert data, and compare throughput performance with other science modules that exist in FINK. The results shown herein emphasise the time-series transformer's suitability for real-time classification at LSST scale, and beyond, and introduce deep model compression as a fundamental tool for improving deploy-ability and scalable inference of deep learning models for transient classification.

For agents

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

get_harvested_code_for_paper("2303.08951")
get_code_for_paper("2303.08951")
have("2303.08951")

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

curl https://syntology.ai/api/ran/2303.08951.json

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

[![Syntology run record](https://syntology.ai/api/ran/2303.08951.svg)](https://syntology.ai/paper/2303.08951)

Connect an agent — have() is free.