Tensor Comprehensions: Framework-Agnostic High-Performance Machine Learning Abstractions · HackerPapers discussions

Tensor Comprehensions: Framework-Agnostic High-Performance Machine Learning Abstractions — mentioned by sidereal

Tue, 22 May 2018 22:08:10 GMT

Polyhedral optimisation is cool (Facebook has been using it to great effect for ML kernels recently [1]), but it’s not the end of the story. It’s complementary to this paper, which seems to be about learning an effective and transferrable cost model to guide the optimisation process (you could use that learned cost model in a polyhedral optimiser). [1]: https://arxiv.org/abs/1802.04730

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

Tensor Comprehensions: Framework-Agnostic High-Performance Machine Learning Abstractions — HN submission

Tue, 20 Feb 2018 09:49:51 GMT

Framework-Agnostic Machine Learning Abstractions (Facebook Research)

3 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Tensor Comprehensions: Framework-Agnostic High-Performance Machine Learning Abstractions — mentioned by ozinenko

Thu, 15 Feb 2018 09:20:19 GMT

Section 7 of the paper (https://arxiv.org/abs/1802.04730) has a couple of examples. In short, yes CuDNN is fast for the cases it was tuned for. It is probably faster on power-of-two sizes, but when you operate on a 26 x 1024954 x 3 tensor, TC can generate specialized code. Want 42 x 17 x 5? TC can generate differently specialized code. With almost no effort from the user (or performance engineers). Can a performance expert do better job than TC optimizer? Very likely yes, but it will very likely take much more time. TC is not a framework. It can be integrated with any framework of your liking.

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