Benchmarking State-of-the-Art Deep Learning Software Tools · HackerPapers discussions

Benchmarking State-of-the-Art Deep Learning Software Tools — mentioned by minimaxir

Fri, 02 Jun 2017 16:27:21 GMT

The important feature is Keras compatability (although it doesn't seem to be in the official repo yet: https://github.com/fchollet/keras/pull/6800) In terms of "how is CNTK better than TensorFlow," CNTK trains 5-10x faster at minimum on pre-2.0 LSTM benchmarks, which is big since LSTMs are used for a lot nowadays. (https://arxiv.org/abs/1608.07249)

Paper and discussion history. A mention is not an endorsement.

Benchmarking State-of-the-Art Deep Learning Software Tools — mentioned by minimaxir

Thu, 01 Jun 2017 16:20:22 GMT

The important feature is Keras compatability: https://docs.microsoft.com/en-us/cognitive-toolkit/Using-CNT... (although it doesn't seem to be in the official repo yet: https://github.com/fchollet/keras/pull/6800) In terms of "how is CNTK better than TensorFlow," CNTK performs 6x better at minimum on old LSTM benchmarks, which is big. (https://arxiv.org/abs/1608.07249)

Paper and discussion history. A mention is not an endorsement.

Benchmarking State-of-the-Art Deep Learning Software Tools — HN submission

Wed, 21 Sep 2016 21:35:04 GMT

New Version of Benchmarking State-Of-the-Art Deep Learning Software Tools

An early version of http://arxiv.org/abs/1608.07249v5 appeared here a month ago. Judging from the changelog (Section 7.1) the authors have fixed many issues. The results differ quite a bit from the original version. On the high end GPUs (GTX 1080) it seems that CNTK is the best for LSTMs and Fully Connected Nets. Torch is the best for Resnet 50 and Caffe is the best for Alexnet.

4 points; 2 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Benchmarking State-of-the-Art Deep Learning Software Tools — HN submission

Sat, 27 Aug 2016 08:04:52 GMT

Benchmarking State-Of-the-Art Deep Learning Software Tools

57 points; 11 discussion replies on HN. Paper and source history. A submission is not an endorsement.