`torch.norm` is very odd. Some notable issues are:
- The default value of `"fro"` in `torch.norm` has an odd behaviour when `dim=None`. This is handled in the new dispatch
- The treatment of the `dtype` argument in `torch.norm` was completely wrong. This should fix it
- Some `out=` variants in the previous implementation were also wrong. This should fix those.
- This new dispatch should make some paths much faster. For example, `torch.norm(x)` where `x` is complex.
I'll try to make the changes in these PRs as incremental as possible as this is a tricky one.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/81761
Approved by: https://github.com/ngimel
As per title. I corrected a thing or two from my previous implementation
to make for better errors in some weird edge-cases and have a more clear
understanding of when does this function support low_precision types and
when it doesn't.
We also use the optimisation for bfloat16 within `vector_norm` within
this function.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/81113
Approved by: https://github.com/ngimel
This PR also adds complex support for logdet, and makes all these
functions support out= and be composite depending on one function. We
also extend the support of `logdet` to complex numbers and improve the
docs of all these functions.
We also use `linalg_lu_factor_ex` in these functions, so we remove the
synchronisation present before.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/79742
Approved by: https://github.com/IvanYashchuk, https://github.com/albanD
This PR heavily simplifies the code of `linalg.solve`. At the same time,
this implementation saves quite a few copies of the input data in some
cases (e.g. A is contiguous)
We also implement it in such a way that the derivative goes from
computing two LU decompositions and two LU solves to no LU
decompositions and one LU solves. It also avoids a number of unnecessary
copies the derivative was unnecessarily performing (at least the copy of
two matrices).
On top of this, we add a `left` kw-only arg that allows the user to
solve `XA = B` rather concisely.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/74046
Approved by: https://github.com/nikitaved, https://github.com/IvanYashchuk, https://github.com/mruberry
This PR simplifies the logic of `linalg.qr` using structured kernels. I
also took this chance and merged a few `copy_` operations with other
ops.
This PR removes a the previous magma implementation as is never faster
than that of cusolver and it's rather buggy. This has the side-effect
that now `qr` is not supported in Rocm. Ivan confirmed that this is
fine, given how incredibly slow was QR on Rocm anyway (we were marking
some tests as slow because of this...).
This PR also corrects the dispatch in geqrf. Before, if we called it
with a matrix for which `input.size(-2) <= 256 && batchCount(input) >= std::max<int64_t>(2, input.size(-2) / 16)` is false, and we have cublas but not cusolver, we would end up calling magma rather than cublas. This is not what the heuristic suggested.
Probaly we should benchmark these heuristics again, but that's beyond the scope of this PR.
Note. It looks like `torch.geqrf` maybe broken in MAGMA as per the
previous comment in `linalg_qr_helper_magma`. IvanYashchuk wdyt?
Pull Request resolved: https://github.com/pytorch/pytorch/pull/79054
Approved by: https://github.com/IvanYashchuk, https://github.com/ezyang
**BC-breaking note**:
This PR deprecates `torch.lu` in favor of `torch.linalg.lu_factor`.
A upgrade guide is added to the documentation for `torch.lu`.
Note this PR DOES NOT remove `torch.lu`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/77636
Approved by: https://github.com/malfet
This PR adds `linalg.lu_solve`. While doing so, I found a bug in MAGMA
when calling the batched MAGMA backend with trans=True. We work around
that by solving the system solving two triangular systems.
We also update the heuristics for this function, as they were fairly
updated. We found that cuSolver is king, so luckily we do not need to
rely on the buggy backend from magma for this function.
We added tests testing this function left and right. We also added tests
for the different backends. We also activated the tests for AMD, as
those should work as well.
Fixes https://github.com/pytorch/pytorch/issues/61657
Pull Request resolved: https://github.com/pytorch/pytorch/pull/77634
Approved by: https://github.com/malfet
This PR does a number of things:
- Move linalg.vector_norm to structured kernels and simplify the logic
- Fixes a number of prexisting issues with the dtype kwarg of these ops
- Heavily simplifies and corrects the logic of `linalg.matrix_norm` and `linalg.norm` to be consistent with the docs
- Before the `_out` versions of these functions were incorrect
- Their implementation is now as efficient as expected, as it avoids reimplementing these operations whenever possible.
- Deprecates `torch.frobenius_norm` and `torch.nuclear_norm`, as they were exposed in the API and they are apparently being used in mobile (??!!) even though they were not documented and their implementation was slow.
- I'd love to get rid of these functions already, but I guess we have to go through their deprecation.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/76547
Approved by: https://github.com/mruberry
This PR adds `linalg.lu_solve`. While doing so, I found a bug in MAGMA
when calling the batched MAGMA backend with trans=True. We work around
that by solving the system solving two triangular systems.
We also update the heuristics for this function, as they were fairly
updated. We found that cuSolver is king, so luckily we do not need to
rely on the buggy backend from magma for this function.
We added tests testing this function left and right. We also added tests
for the different backends. We also activated the tests for AMD, as
those should work as well.
Fixes https://github.com/pytorch/pytorch/issues/61657
Pull Request resolved: https://github.com/pytorch/pytorch/pull/72935
Approved by: https://github.com/IvanYashchuk, https://github.com/mruberry
This PR modifies `lu_unpack` by:
- Using less memory when unpacking `L` and `U`
- Fuse the subtraction by `-1` with `unpack_pivots_stub`
- Define tensors of the correct types to avoid copies
- Port `lu_unpack` to be a strucutred kernel so that its `_out` version
does not incur on extra copies
Then we implement `linalg.lu` as a structured kernel, as we want to
compute its derivative manually. We do so because composing the
derivatives of `torch.lu_factor` and `torch.lu_unpack` would be less efficient.
This new function and `lu_unpack` comes with all the things it can come:
forward and backward ad, decent docs, correctness tests, OpInfo, complex support,
support for metatensors and support for vmap and vmap over the gradients.
I really hope we don't continue adding more features.
This PR also avoids saving some of the tensors that were previously
saved unnecessarily for the backward in `lu_factor_ex_backward` and
`lu_backward` and does some other general improvements here and there
to the forward and backward AD formulae of other related functions.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/67833
Approved by: https://github.com/IvanYashchuk, https://github.com/nikitaved, https://github.com/mruberry
We derive and implement a more concise rule for the forward and backward
derivatives of the QR decomposition. While doing this we:
- Fix the composite compliance of `linalg.qr` and we make it support batches
- Improve the performance and simplify the implementation of both foward and backward
- Avoid saving the input matrix for the backward computation.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/76115
Approved by: https://github.com/nikitaved, https://github.com/albanD
Let's make sure we don't break anything in the next PRs of the stack.
Also some comprehensive testing of matmul on CPU and CUDA was long due.
Running this tests we see that the `out=` variant of matmul is broken
when used on 4D tensors. This hints what would be the amount of people
that use out= variants...
Pull Request resolved: https://github.com/pytorch/pytorch/pull/75193
Approved by: https://github.com/ngimel
This PR adds a function for computing the LDL decomposition and a function that can solve systems of linear equations using this decomposition. The result of `torch.linalg.ldl_factor_ex` is in a compact form and it's required to use it only through `torch.linalg.ldl_solve`. In the future, we could provide `ldl_unpack` function that transforms the compact representation into explicit matrices.
Fixes https://github.com/pytorch/pytorch/issues/54847.
cc @jianyuh @nikitaved @pearu @mruberry @walterddr @IvanYashchuk @xwang233 @Lezcano
Pull Request resolved: https://github.com/pytorch/pytorch/pull/69828
Approved by: https://github.com/Lezcano, https://github.com/mruberry, https://github.com/albanD
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/73748
This adds CPU-only slow test jobs, which previously would never run.
Includes fixes/skips for slow tests which fail (they need to be skipped now because they used to never run)
Test Plan: Imported from OSS
Reviewed By: malfet
Differential Revision: D34628803
Pulled By: davidberard98
fbshipit-source-id: c090ab7bf7bda9e24ec5cdefa6fd35c6310dbac0
(cherry picked from commit 06f7a94a57cc7023e9c5442be8298d20cd011144)
