TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
Citations
13,268 citations
Cites methods from "TensorFlow: Large-Scale Machine Lea..."
...Dynamic execution distinguishes PyTorch from static frameworks like TensorFlow [1], Caffe, etc. • Immediate, eager execution....
[...]
...Dynamic execution distinguishes PyTorch from static frameworks like TensorFlow [1], Caffe, etc....
[...]
11,856 citations
Cites methods from "TensorFlow: Large-Scale Machine Lea..."
...We have implemented the second approach into the TensorFlow framework [83]....
[...]
10,913 citations
10,601 citations
10,120 citations
References
72,897 citations
"TensorFlow: Large-Scale Machine Lea..." refers methods in this paper
...ng hand-written digits from the MNIST dataset (the “hello world” of machine learning algorithms) [32], classifying images from the CIFAR10 dataset [30], doing language modeling using a recurrent LSTM [22] network, training word embedding vectors [35] and more. The system includes front-ends for specifying TensorFlow computations in Python and C++, and we expect other front-ends to be added over time i...
[...]
40,257 citations
30,843 citations
"TensorFlow: Large-Scale Machine Lea..." refers background in this paper
...In particular, we focus on our lessons from porting a state-of-the-art convolutional neural network for image recognition termed Inception [23]....
[...]
...Given these circumstances, we found the following strategies critical for porting the Inception model to TensorFlow: 1....
[...]
...The strategies outlined above proved invaluable in gaining confidence in the system and ultimately in instantiating the Inception model in TensorFlow....
[...]
23,814 citations
20,077 citations