Keras: The Python Deep Learning library

By François Chollet. 2018

1,498 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/77cec64cdf0581ff4e7b70aa595a4d6e39b70ce4

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Efficiently Identifying Task Groupings for Multi-Task Learning (2021, methodology): “Our experiments were run on a combination of Keras [12] and TensorFlow [1].”
Ares: A framework for quantifying the resilience of deep neural networks (2018, methodology): “Ares is built on top of Keras [11], which takes high-level DNN descriptions specified in Python and executes them using either Theano [12] or TensorFlow [13] back ends.”
Kvasir-SEG: A Segmented Polyp Dataset (2019, methodology): “For our deep-learning implementations, we used the Keras framework [8] and Tensorflow [1] as a backend.”
A fully integrated computer-aided diagnosis system for digital X-ray mammograms via deep learning detection, segmentation, and classification (2018, methodology): “The implementation of all deep segmentation models is achieved utilizing Theano [62] and Keras [63] deep learning libraries, while the detection and classi ﬁ cation models are implemented under the Tensor ﬂ ow environment [64].”
An extensive experimental survey of regression methods (2019, methodology): “dlkeras is the deep learning neural network implemented by the Keras 1262 module [67] of the Python programming language, with three hidden 1263 layers tuned with 50 and 75 neurons for each layer (nh1, nh2 and nh3, 1264 with 8 combinations).”
Online capacity estimation of lithium-ion batteries with deep long short-term memory networks (2021, methodology): “In this paper, the LSTM-RNN architecture was developed in a Python 3 environment, with TensorFlow 1.14 as the backend and the Keras deep learning library used for layer creation [45,46].”
Detecting Cyber Attacks in Industrial Control Systems Using Convolutional Neural Networks (2018, methodology): “ framework version 1.4 [9]. Specifically, we used the low-level API which allowed for finegrained control over the network architecture. We also evaluated an implementation based on the Keras library [3], but found that TensorFlow provided slightly faster training and test times. 6.2 Data Preprocess”
Crepe: A Convolutional Representation for Pitch Estimation (2018, methodology): “This architecture and the training procedures are implemented using Keras [23].”