DASH: Deep Learning for the Automated Spectral Classification of Supernovae and Their Hosts (2024)

Muthukrishna, D., Parkinson, D., & Tucker, B. E. (2019). DASH: Deep Learning for the Automated Spectral Classification of Supernovae and Their Hosts. Astrophysical Journal, 885(1), Article 85. https://doi.org/10.3847/1538-4357/ab48f4

Muthukrishna, Daniel ; Parkinson, David ; Tucker, Brad E. / DASH : Deep Learning for the Automated Spectral Classification of Supernovae and Their Hosts. In: Astrophysical Journal. 2019 ; Vol. 885, No. 1.

@article{b489190d68d24a50a5b25d45e7e93e0a,

title = "DASH: Deep Learning for the Automated Spectral Classification of Supernovae and Their Hosts",

abstract = "We present DASH (Deep Automated Supernova and Host classifier), a novel software package that automates the classification of the type, age, redshift, and host galaxy of supernova spectra. DASH makes use of a new approach that does not rely on iterative template-matching techniques like all previous software, but instead classifies based on the learned features of each supernova's type and age. It has achieved this by employing a deep convolutional neural network to train a matching algorithm. This approach has enabled DASH to be orders of magnitude faster than previous tools, being able to accurately classify hundreds or thousands of objects within seconds. We have tested its performance on 4 yr of data from the Australian Dark Energy Survey (OzDES). The deep learning models were developed using TensorFlow and were trained using over 4000 supernova spectra taken from the CfA Supernova Program and the Berkeley SN Ia Program as used in SNID (Supernova Identification software). Unlike template-matching methods, the trained models are independent of the number of spectra in the training data, which allows for DASH's unprecedented speed. We have developed both a graphical interface for easy visual classification and analysis of supernovae and a Python library for the autonomous and quick classification of several supernova spectra. The speed, accuracy, user-friendliness, and versatility of DASH present an advancement to existing spectral classification tools. We have made the code publicly available on GitHub and PyPI (pip install astrodash) to allow for further contributions and development. The package documentation is available at https://astrodash.readthedocs.io.",

keywords = "methods: data analysis, methods: statistical, supernovae: general, surveys, techniques: spectroscopic",

author = "Daniel Muthukrishna and David Parkinson and Tucker, {Brad E.}",

note = "Publisher Copyright: {\textcopyright} 2019. The American Astronomical Society. All rights reserved.",

year = "2019",

month = nov,

day = "1",

doi = "10.3847/1538-4357/ab48f4",

language = "English",

volume = "885",

journal = "Astrophysical Journal",

issn = "0004-637X",

publisher = "American Astronomical Society",

number = "1",

}

Muthukrishna, D, Parkinson, D & Tucker, BE 2019, 'DASH: Deep Learning for the Automated Spectral Classification of Supernovae and Their Hosts', Astrophysical Journal, vol. 885, no. 1, 85. https://doi.org/10.3847/1538-4357/ab48f4

DASH: Deep Learning for the Automated Spectral Classification of Supernovae and Their Hosts. / Muthukrishna, Daniel; Parkinson, David; Tucker, Brad E.
In: Astrophysical Journal, Vol. 885, No. 1, 85, 01.11.2019.

Research output: Contribution to journal › Article › peer-review

TY - JOUR

T1 - DASH

T2 - Deep Learning for the Automated Spectral Classification of Supernovae and Their Hosts

AU - Muthukrishna, Daniel

AU - Parkinson, David

AU - Tucker, Brad E.

N1 - Publisher Copyright:© 2019. The American Astronomical Society. All rights reserved.

PY - 2019/11/1

Y1 - 2019/11/1

N2 - We present DASH (Deep Automated Supernova and Host classifier), a novel software package that automates the classification of the type, age, redshift, and host galaxy of supernova spectra. DASH makes use of a new approach that does not rely on iterative template-matching techniques like all previous software, but instead classifies based on the learned features of each supernova's type and age. It has achieved this by employing a deep convolutional neural network to train a matching algorithm. This approach has enabled DASH to be orders of magnitude faster than previous tools, being able to accurately classify hundreds or thousands of objects within seconds. We have tested its performance on 4 yr of data from the Australian Dark Energy Survey (OzDES). The deep learning models were developed using TensorFlow and were trained using over 4000 supernova spectra taken from the CfA Supernova Program and the Berkeley SN Ia Program as used in SNID (Supernova Identification software). Unlike template-matching methods, the trained models are independent of the number of spectra in the training data, which allows for DASH's unprecedented speed. We have developed both a graphical interface for easy visual classification and analysis of supernovae and a Python library for the autonomous and quick classification of several supernova spectra. The speed, accuracy, user-friendliness, and versatility of DASH present an advancement to existing spectral classification tools. We have made the code publicly available on GitHub and PyPI (pip install astrodash) to allow for further contributions and development. The package documentation is available at https://astrodash.readthedocs.io.

AB - We present DASH (Deep Automated Supernova and Host classifier), a novel software package that automates the classification of the type, age, redshift, and host galaxy of supernova spectra. DASH makes use of a new approach that does not rely on iterative template-matching techniques like all previous software, but instead classifies based on the learned features of each supernova's type and age. It has achieved this by employing a deep convolutional neural network to train a matching algorithm. This approach has enabled DASH to be orders of magnitude faster than previous tools, being able to accurately classify hundreds or thousands of objects within seconds. We have tested its performance on 4 yr of data from the Australian Dark Energy Survey (OzDES). The deep learning models were developed using TensorFlow and were trained using over 4000 supernova spectra taken from the CfA Supernova Program and the Berkeley SN Ia Program as used in SNID (Supernova Identification software). Unlike template-matching methods, the trained models are independent of the number of spectra in the training data, which allows for DASH's unprecedented speed. We have developed both a graphical interface for easy visual classification and analysis of supernovae and a Python library for the autonomous and quick classification of several supernova spectra. The speed, accuracy, user-friendliness, and versatility of DASH present an advancement to existing spectral classification tools. We have made the code publicly available on GitHub and PyPI (pip install astrodash) to allow for further contributions and development. The package documentation is available at https://astrodash.readthedocs.io.

KW - methods: data analysis

KW - methods: statistical

KW - supernovae: general

KW - surveys

KW - techniques: spectroscopic

UR - http://www.scopus.com/inward/record.url?scp=85075261449&partnerID=8YFLogxK

U2 - 10.3847/1538-4357/ab48f4

DO - 10.3847/1538-4357/ab48f4

M3 - Article

SN - 0004-637X

VL - 885

JO - Astrophysical Journal

JF - Astrophysical Journal

IS - 1

M1 - 85

ER -

Muthukrishna D, Parkinson D, Tucker BE. DASH: Deep Learning for the Automated Spectral Classification of Supernovae and Their Hosts. Astrophysical Journal. 2019 Nov 1;885(1):85. doi: 10.3847/1538-4357/ab48f4