43 Hits in 4.2 sec

Comparative Study of Deep Learning Software Frameworks [article]

Soheil Bahrampour, Naveen Ramakrishnan, Lukas Schott, Mohak Shah
2016 arXiv   pre-print
This paper presents a comparative study of five deep learning frameworks, namely Caffe, Neon, TensorFlow, Theano, and Torch, on three aspects: extensibility, hardware utilization, and speed.  ...  Theano achieves the best performance on GPU for training and deployment of LSTM networks. Caffe is the easiest for evaluating the performance of standard deep architectures.  ...  of the top deep learning frameworks, namely Caffe, Neon, TensorFlow, Theano and Torch for a variety of settings on a single machine.  ... 
arXiv:1511.06435v3 fatcat:6em43mvhbjdv3hheg3y7e466y4

A detailed comparative study of open source deep learning frameworks [article]

Ghadeer Al-Bdour, Raffi Al-Qurran, Mahmoud Al-Ayyoub, Ali Shatnawi
2020 arXiv   pre-print
Deep Learning (DL) is one of the hottest trends in machine learning as DL approaches produced results superior to the state-of-the-art in problematic areas such as image processing and natural language  ...  Theano and Microsoft's CNTK).  ...  The authors compared five DL frameworks: Caffe, Neon, TensorFlow, Theano, and Torch, in terms of speed (gradient computation time and forward time), hardware utilization and extensibility (ability to support  ... 
arXiv:1903.00102v2 fatcat:wzaa3ux7tndk3gzd235ki2773y

A Comparison of the State-of-the-Art Deep Learning Platforms: An Experimental Study

Abdullah Talha KABAKUŞ
2020 Sakarya University Journal of Computer and Information Sciences  
This experimental study should help deep learning engineers and researchers to choose the most suitable platform for the implementations of their deep neural networks.  ...  Two state-of-the-art deep learning platforms, namely, ( ) Keras, and ( ) PyTorch were included in the comparison within this study.  ...  [23] proposed a comparative study of Caffe, neon [24] , Theano, and Torch for deep learning tasks.  ... 
doi:10.35377/saucis.03.03.776573 fatcat:fbiso2ournh47fbvsej3svgwtm

Benchmarking open source deep learning frameworks

Ghadeer Al-Bdour, Raffi Al-Qurran, Mahmoud Al-Ayyoub, Ali Shatnawi
2020 International Journal of Electrical and Computer Engineering (IJECE)  
Deep Learning (DL) is one of the hottest fields. To foster the growth of DL, several open source frameworks appeared providing implementations of the most common DL algorithms.  ...  The purpose of this work is to provide a qualitative and quantitative comparison among three such frameworks: TensorFlow, Theano and CNTK.  ...  Besides, they used three NN types: CNN, AutoEncoder (AE) and LSTM to train MNIST, ImageNet [25] and IMDB datasets on Caffe, Neon, TensorFlow, Theano and Torch frameworks.  ... 
doi:10.11591/ijece.v10i5.pp5479-5486 fatcat:ypbgo5yhybhf5hfj2uoqn7qksq

A Survey on Deep Learning Architectures and Frameworks for Cancer Detection in Medical Images Analysis

The paper presents the study of a few deep learning software frameworks such as tensor flow, theano, caffe, torch, and keras. Tensor Flow provides excellent functionality for deep learning.  ...  The study includes some dominant deep learning algorithms such as convolution neural network, fully convolutional network, autoencoder, and deep belief network to analyze the medical image and to detect  ...  TensorFlow, Theano, Caffe, Torch, and Keras, the five deep learning frameworks are analyzed to provide a study in it.  ... 
doi:10.35940/ijitee.k7654.0991120 fatcat:4qkd3kdqvjgu7n2g7wnmnyiici

A Step Forward Towards Deep Learning

Varshapriya J., Minal Ugale
2017 International Journal of Advanced Research in Computer Science and Software Engineering  
Also, the demand for a machine engineer or a deep learning scientist has increased manifold. With this rate, knowing and applying deep learning would not only remain a skill but a necessity.  ...  The aim of this survey paper is to make one familiar with the fundamentals of deep learning and how to take a step forward from knowing the buzzword to knowing the subject.  ...  The list of available frameworks includes, but is not limited to, Caffe, DeepLearning4J, deepmat, Eblearn, Neon, PyLearn, TensorFlow, Theano, Torch, etc.  ... 
doi:10.23956/ijarcsse/v7i3/0121 fatcat:kkww74cx4ffmjltcyqcqx4rrlq

Deep Learning in Mining Biological Data [article]

Mufti Mahmud, M Shamim Kaiser, Amir Hussain
2020 arXiv   pre-print
Artificial neural network-based learning systems are well known for their pattern recognition capabilities and lately their deep architectures - known as deep learning (DL) - have been successfully applied  ...  Mining such an enormous amount of data for pattern recognition is a big challenge and requires sophisticated data-intensive machine learning techniques.  ...  Besides, different DL tools have different targets, e.g., Caffe aims applications, whereas, Torch and Theano are more for DL research.  ... 
arXiv:2003.00108v1 fatcat:vd5yc3dvl5bvboxfkytja4zlyy

Performance Analysis of Deep Learning Libraries: TensorFlow and PyTorch

Felipe Florencio, Thiago Valen�, Edward David Moreno, Methanias Cola�o Junior
2019 Journal of Computer Science  
Through the increase in deep learning study and use, in the last years there was a development of specific libraries for Deep Neural Network (DNN).  ...  For this reason, developers and scientists that work with deep learning need scientific experimental studies that examine the performance of those libraries.  ...  Edward David Moreno: Coordinated the experiments and contributed to the writing of the manuscript. Dr. Methanias Colaço Junior: Designed the research plan, organized the study and contributed.  ... 
doi:10.3844/jcssp.2019.785.799 fatcat:7utblu4dtfevfdrwdv76nl6cyq

DLBench: a comprehensive experimental evaluation of deep learning frameworks

Radwa Elshawi, Abdul Wahab, Ahmed Barnawi, Sherif Sakr
2021 Cluster Computing  
In practice, the increasing popularity of deep learning frameworks calls for benchmarking studies that can effectively evaluate and understand the performance characteristics of these systems.  ...  (e.g., GPU) and the growing number of open source deep learning frameworks that facilitate and ease the development process of deep learning architectures.  ...  The authors would like to thank the students Nesma Mahmoud, Yousef Essam, and Hassan Eldeeb for their involvement on some of the experiments of this work.  ... 
doi:10.1007/s10586-021-03240-4 fatcat:eycfx5pmtvclhnhkwof2zu6nsa

Deep Learning in Data-Driven Pavement Image Analysis and Automated Distress Detection: A Review

Kasthurirangan Gopalakrishnan
2018 Data  
The review concludes with potential avenues for future research; especially in the application of deep learning to not only detect, but also characterize the type, extent, and severity of distresses from  ...  A comparison of the deep learning software frameworks, network architecture, hyper-parameters employed by each study, and crack detection performance is provided, which is expected to provide a good foundation  ...  Conflicts of Interest: The author declares no conflicts of interest.  ... 
doi:10.3390/data3030028 fatcat:u2g4mniiajgz7ohpeylsrsrsvm


Moustafa Alzantot, Yingnan Wang, Zhengshuang Ren, Mani B. Srivastava
2017 Proceedings of the 1st International Workshop on Deep Learning for Mobile Systems and Applications - EMDL '17  
In this paper, we study the benefits of utilizing the heterogeneous (CPU and GPU) computing resources available on commodity android devices while running deep learning models.  ...  However, popular and commonly used tools and frameworks for machine intelligence are still lacking the ability to make proper use of the available heterogeneous computing resources on mobile devices.  ...  Acknowledgments This research was supported in part by the NIH Center of Excellence for Mobile Sensor Data-to-Knowledge (MD2K) under award 1-U54EB020404-01, and by the U.S.  ... 
doi:10.1145/3089801.3089805 pmid:29629431 pmcid:PMC5889131 dblp:conf/mobisys/AlzantotWRS17 fatcat:j3hd5ruiu5hshd2qs2icfvj32q

Deep Learning in Mining Biological Data

Mufti Mahmud, M. Shamim Kaiser, T. Martin McGinnity, Amir Hussain
2021 Cognitive Computation  
Artificial neural network-based learning systems are well known for their pattern recognition capabilities, and lately their deep architectures—known as deep learning (DL)—have been successfully applied  ...  Mining such enormous amount of data for pattern recognition is a big challenge and requires sophisticated data-intensive machine learning techniques.  ...  Acknowledgements The authors would like to thank the members of the acslab (http://www.acsla for valuable discussions. Author Contributions  ... 
doi:10.1007/s12559-020-09773-x pmid:33425045 pmcid:PMC7783296 fatcat:n4nk7gakfbb4fbhdi5pqeojwjm

A Survey on Deep Learning Methods for Robot Vision [article]

Javier Ruiz-del-Solar, Patricio Loncomilla, Naiomi Soto
2018 arXiv   pre-print
Then, the standard methodology and tools used for designing deep-learning based vision systems are presented.  ...  Afterwards, a review of the principal work using deep learning in robot vision is presented, as well as current and future trends related to the use of deep learning in robotics.  ...  [ 170 ] 170 five popular frameworks are compared (Caffe, Neon, Tensorflow, Theano, and Torch).  ... 
arXiv:1803.10862v1 fatcat:bkxbwfkuxbfkrck4aafdgt3moy

Performance analysis of real-time DNN inference on Raspberry Pi

Delia Velasco-Montero, Jorge Fernández-Berni, Ricardo Carmona-Galán, Ángel Rodríguez-Vázquez, Nasser Kehtarnavaz, Matthias F. Carlsohn
2018 Real-Time Image and Video Processing 2018  
Caffe, OpenCV, TensorFlow, Theano, Torch or MXNet.  ...  In this paper, we present a comparative study of some of these frameworks in terms of power consumption, throughput and precision for some of the most popular Convolutional Neural Networks (CNN) models  ...  Enrique Pardo for his help to set up benchmarking components.  ... 
doi:10.1117/12.2309763 fatcat:kts32uab4nbqrndnhgkdknadq4

Towards Detecting Dementia via Deep Learning

Deepika Bansal, *Kavita Khanna, Rita Chhikara, Rakesh Kumar Dua, Rajeev Malini
2021 International Journal of Healthcare Information Systems and Informatics  
Deep learning provides path-breaking applications in medical imaging. This study provides a detailed summary of different implementation approaches of deep learning for detecting the disease.  ...  Accuracy of the methods has been compared and the best parameters including classifier, learning rate, and a batch size of the model have been identified.  ...  ACKNowLeDGMeNT The research was funded by the Department of Science and Technology DST, New Delhi, Reference number DST/CSRI/2017/215 (G).  ... 
doi:10.4018/ijhisi.20211001.oa31 fatcat:yjxrrsej2jgahjxrm37msopbkm
« Previous Showing results 1 — 15 out of 43 results