missing algebra gradients

[CarloLucibello]

• [x] diag, diagm
• [x] triu, tril
• [x] kron
• [x] dot
• [x] inv
• [x] trace
• [x] det (see https://github.com/pytorch/pytorch/pull/3816)
• [x] logdet, logabsdet
• [x] chol (see https://github.com/tensorflow/tensorflow/pull/1465 and the paper mentioned over there)
• [x] svd (see https://github.com/pytorch/pytorch/pull/3816 and https://j-towns.github.io/papers/svd-derivative.pdf )
• [x] qr (see https://github.com/tensorflow/tensorflow/issues/6504#issuecomment-316892961)
• [x] lq
• [ ] eig
• [ ] eigs
• [ ] \, /

useful reading https://arxiv.org/pdf/1710.08717.pdf

[denizyuret]

All done except last 3?

[CarloLucibello]

All done except last 3?

right

[denizyuret]

Some of these broke because of 0.7. I put @test_broken in test/linearalgebra.jl