OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-07 12:21:27 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity
64,482 Commits 8,304 Branches 150 Tags 1.5 GiB
1b51d29b66
Commit Graph

421 Commits

Author SHA1 Message Date
Nikita Karetnikov
e9705c52ac [pt2] add metas for _pdist_forward and _pdist_backward (#103817) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103817
Approved by: https://github.com/ezyang
 2023-06-22 11:18:05 +00:00
Nikita Karetnikov
e48851033a [pt2] add metas for pad ops (#103815) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103815
Approved by: https://github.com/ezyang
 2023-06-22 11:18:05 +00:00
Brian Hirsh
c3c03e7cb8 Reland of https://github.com/pytorch/pytorch/pull/101818 (#103888) 
Original PR broke internal

This reverts commit 5ed618132f.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/103888
Approved by: https://github.com/albanD
 2023-06-21 21:00:56 +00:00
Peter Bell
8b418f197c [decomp] Add decomposition for torch.renorm (#103858) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103858
Approved by: https://github.com/ezyang, https://github.com/nkaretnikov
 2023-06-21 20:57:43 +00:00
Peter Bell
a61096fb94 [decomp] Decompose logaddexp2 (#103765) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103765
Approved by: https://github.com/Chillee
 2023-06-21 20:16:24 +00:00
PyTorch MergeBot
7b6dc72ffa Revert "[decomp] Decompose logaddexp2 (#103765)" 
This reverts commit bab21d20eb.

Reverted https://github.com/pytorch/pytorch/pull/103765 on behalf of https://github.com/ezyang due to looks like land race ([comment](https://github.com/pytorch/pytorch/pull/103765#issuecomment-1599030496))
 2023-06-20 15:35:02 +00:00
Peter Bell
bab21d20eb [decomp] Decompose logaddexp2 (#103765) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103765
Approved by: https://github.com/Chillee
 2023-06-20 09:24:21 +00:00
Richard Zou
27a67d8699 Refactor and improve make_fx testing (#103196) 
This is in preparation for the custom_op_compile_check utility, which
will call the newly refactored function.

This PR:
- splits off code into helper functions
- adds clearer error messages
- stops updating the inputs destructively (leading to slightly slower
tests)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103196
Approved by: https://github.com/bdhirsh, https://github.com/soulitzer
 2023-06-14 14:00:12 +00:00
Nikita Karetnikov
d38b651d51 [pt2] add SymInt support for cosine_similarity (#103400) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103400
Approved by: https://github.com/ezyang, https://github.com/Skylion007
 2023-06-13 21:23:48 +00:00
Nikita Karetnikov
c07634436e [pt2] add SymInt support for bilinear (#103396) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103396
Approved by: https://github.com/ezyang
 2023-06-13 21:23:48 +00:00
Nikita Karetnikov
4a76fb49f3 [pt2] add metas for avg_pool3d and avg_pool3d_backward (#103392) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103392
Approved by: https://github.com/ezyang
 2023-06-13 21:23:46 +00:00
PyTorch MergeBot
5ed618132f Revert "change pre_autograd to pre_dispatch tracing (#101818)" 
This reverts commit b0392de2c3.

Reverted https://github.com/pytorch/pytorch/pull/101818 on behalf of https://github.com/izaitsevfb due to Breaks internal builds see D46629736 TypeError: wrap_key() got an unexpected keyword argument pre_autograd ([comment](https://github.com/pytorch/pytorch/pull/101818#issuecomment-1587837667))
 2023-06-12 18:16:37 +00:00
Nikita Karetnikov
2b3d955ffd [pt2] add meta and SymInt support for linalg_matrix_exp (#102945) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102945
Approved by: https://github.com/lezcano
 2023-06-09 22:45:16 +00:00
Brian Hirsh
b0392de2c3 change pre_autograd to pre_dispatch tracing (#101818) 
We discussed in a composability meeting a few weeks ago that `pre_autograd` should probably be renamed to `pre_dispatch`.

One question in this PR was: should I re-use a dispatch key? Or should I create a new dispatch key (that yet again corresponds to "top of the dispatcher")?

~~For now, I ended up sticking our proxy mode on the mode stack corresponding to `PythonTLSSnapshot`, because it was simple and it works. It looks like one of the functorch dispatch keys has higher priority though, so it's possible that functorch will end up running first. Open to options, but we can consider adding a new dispatch key later if that becomes a problem~~

Update: I added a dedicated dispatch key, `PreDispatch`.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/101818
Approved by: https://github.com/ezyang, https://github.com/Neilblaze, https://github.com/albanD, https://github.com/zou3519
 2023-06-09 17:30:15 +00:00
Nikita Karetnikov
1fcc67fd8c [pt2] add SymInt support for linalg.tensorsolve (#102466) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102466
Approved by: https://github.com/Skylion007, https://github.com/lezcano
 2023-06-06 08:06:55 +00:00
Nikita Karetnikov
ec0aa965da [pt2] add meta for _linalg_solve_ex (#102454) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102454
Approved by: https://github.com/lezcano
 2023-06-06 08:06:55 +00:00
Nikita Karetnikov
4bda4a7e4d [pt2] add meta for lu_unpack (#102937) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102937
Approved by: https://github.com/lezcano
 2023-06-06 08:06:53 +00:00
Nikita Karetnikov
6ac3352a37 [pt2] add meta for _linalg_slogdet (#102464) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102464
Approved by: https://github.com/ezyang
 2023-06-05 03:17:08 +00:00
Nikita Karetnikov
757791d1e3 [pt2] add SymInt support for linalg.vander (#102469) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102469
Approved by: https://github.com/Skylion007, https://github.com/lezcano
 2023-06-04 09:58:02 +00:00
Edward Z. Yang
8bbef821c3 Add some unit tests from cm3leon involving repeat_interleave (#102733) 
These actually were fixed by https://github.com/pytorch/pytorch/pull/102570
but that PR doesn't test guard-freeness, so here you go.

Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102733
Approved by: https://github.com/zou3519
 2023-06-02 15:35:35 +00:00
PyTorch MergeBot
95cdd58c8f Revert "[pt2] add SymInt support for linalg.tensorsolve (#102466)" 
This reverts commit b1b76f614d.

Reverted https://github.com/pytorch/pytorch/pull/102466 on behalf of https://github.com/clee2000 due to reverting b/c stack https://github.com/pytorch/pytorch/pull/102469#issuecomment-1569041604, i think this is the one that actually causes the test to fail ([comment](https://github.com/pytorch/pytorch/pull/102466#issuecomment-1569045123))
 2023-05-30 20:26:46 +00:00
PyTorch MergeBot
463df86ce8 Revert "[pt2] add SymInt support for linalg.vander (#102469)" 
This reverts commit 05717895aa.

Reverted https://github.com/pytorch/pytorch/pull/102469 on behalf of https://github.com/clee2000 due to broke test_aotdispatch on linux ex 05717895aa https://github.com/pytorch/pytorch/actions/runs/5125654882/jobs/9219389448, shows up as green on pr due to bug with keep-going flag and reruns ([comment](https://github.com/pytorch/pytorch/pull/102469#issuecomment-1569041604))
 2023-05-30 20:24:26 +00:00
Nikita Karetnikov
05717895aa [pt2] add SymInt support for linalg.vander (#102469) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102469
Approved by: https://github.com/Skylion007, https://github.com/lezcano
 2023-05-30 19:50:16 +00:00
Nikita Karetnikov
b1b76f614d [pt2] add SymInt support for linalg.tensorsolve (#102466) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102466
Approved by: https://github.com/Skylion007, https://github.com/lezcano
 2023-05-30 19:50:15 +00:00
Nikita Karetnikov
0ba81ce8fe [pt2] add SymInt support for linalg.tensorinv (#102465) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102465
Approved by: https://github.com/Skylion007, https://github.com/lezcano
 2023-05-30 19:50:14 +00:00
Nikita Karetnikov
995ac703cd [pt2] add SymInt support for linalg.pinv (#102367) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102367
Approved by: https://github.com/lezcano
 2023-05-27 11:10:47 +00:00
PyTorch MergeBot
da3aba1e46 Revert "[pt2] add SymInt support for linalg.pinv (#102367)" 
This reverts commit 0d5b74da0c.

Reverted https://github.com/pytorch/pytorch/pull/102367 on behalf of https://github.com/kit1980 due to Broke slow tests https://github.com/pytorch/pytorch/actions/runs/5095190248/jobs/9160028124 ([comment](https://github.com/pytorch/pytorch/pull/102367#issuecomment-1565104562))
 2023-05-27 00:33:42 +00:00
Nikita Karetnikov
0d5b74da0c [pt2] add SymInt support for linalg.pinv (#102367) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102367
Approved by: https://github.com/lezcano
 2023-05-26 15:20:34 +00:00
vfdev-5
e3d97b6213 [inductor] Added smooth_l1_loss refs (#102077) 
Added `smooth_l1_loss` to refs + tests

Pull Request resolved: https://github.com/pytorch/pytorch/pull/102077
Approved by: https://github.com/lezcano, https://github.com/ngimel
 2023-05-24 15:07:08 +00:00
Edward Z. Yang
3318a832b3 Tighten FakeTensor reentrancy asserts, add debugging (#102091) 
When investigating failures in https://github.com/pytorch/pytorch/pull/100017 I realized that we were reentering FakeTensorMode even though there was already one on the stack. Although we have attempted assert for these cases in the past, e.g., as in https://github.com/pytorch/pytorch/pull/97186 it seems that the existing protections were insufficient.

In this particular case, the reapplication of FakeTensorMode was due to an interaction with NotImplemented multiple dispatch handling. If proxy tensor mode detects an unrecognized tensor type (this includes FakeTensor, if it is not tracked with a proxy), it will return NotImplemented to give this tensor a chance to unpack itself into proxyable operation. However, this is never the right thing for FakeTensor, where no unpacking is possible. However, today, FakeTensor attempts to reapply the FakeTensorMode, resulting in FakeTensorMode being twice on the stack.

This PR does a number of things:

* It adds an assert in `FakeTensorMode.__torch_dispatch__` that you must not already have this mode on the stack, this is ALWAYS an error
* It modifies `FakeTensor.__torch_dispatch__` to return `NotImplemented` if the mode is already active. This prevents us from readding the mode on the stack
* It adds a new logging artifact `not_implemented` which you can use to get debug logs about all of the times a `__torch_dispatch__` handler returned NotImplemented and why it did so. Your subclass has to manually opt into this logging, but I inserted the necessary logs for ProxyTensorMode and FakeTensor(Mode)
* `with fake_mode` now no-ops if the fake mode is already on the stack, which is what users want anyway
* I am BREAKING pre-autograd tracing, because it is currently doing something weird with the original C++ mode stack. Brian is going to follow up with a fix next week.

Signed-off-by: Edward Z. Yang <ezyang@meta.com>

Pull Request resolved: https://github.com/pytorch/pytorch/pull/102091
Approved by: https://github.com/thiagocrepaldi, https://github.com/eellison, https://github.com/wanchaol, https://github.com/bdhirsh
 2023-05-24 05:37:51 +00:00
Nikita Karetnikov
e79d9b9938 [pt2] add SymInt support for linalg.matrix_power (#101940) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/101940
Approved by: https://github.com/lezcano, https://github.com/ezyang
 2023-05-24 00:21:52 +00:00
Nikita Karetnikov
42b974e8f7 [pt2] add meta for linalg_lu_solve (#101836) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/101836
Approved by: https://github.com/lezcano
 2023-05-24 00:21:50 +00:00
Khushi
1aaf0396eb [reland][opinfo] empty_strided (#101782) 
Follows #100223

Previous PR: #100890

Pull Request resolved: https://github.com/pytorch/pytorch/pull/101782
Approved by: https://github.com/ezyang
 2023-05-19 03:06:29 +00:00
drisspg
6f13d6892a Add meta support for multinomial (#101324) 
# Summary
Found this when trying to compile the text gen loop of nanogpt here: b33289942b/torchbenchmark/models/nanogpt_generate/model.py (L322)

Pull Request resolved: https://github.com/pytorch/pytorch/pull/101324
Approved by: https://github.com/ngimel
 2023-05-19 00:04:26 +00:00
Angela Yi
72a73ef67b Add aten.searchsorted.Tensor meta kernel (#101637) 
Test Plan: CI

Differential Revision: D45933187

Pull Request resolved: https://github.com/pytorch/pytorch/pull/101637
Approved by: https://github.com/ezyang
 2023-05-18 06:55:11 +00:00
PyTorch MergeBot
dfac4364c4 Revert "[opinfo] empty_strided (#100890)" 
This reverts commit 01c7106580.

Reverted https://github.com/pytorch/pytorch/pull/100890 on behalf of https://github.com/PaliC due to broke test_ops.py slow test ([comment](https://github.com/pytorch/pytorch/pull/100890#issuecomment-1551903975))
 2023-05-17 19:00:15 +00:00
ydwu4
326a4cc815 Support map autograd and pytree in/out. (#101633) 
Rebased https://github.com/pytorch/pytorch/pull/100494 and added dummy AOTConfig.

This PR adds autograd and pytree support for map operator.

Implementation-wise:

1. We temporarily make two HigherOrderOperators, "map" and "map_impl":
- "map" is user-facing. Currently, it unwraps the pytrees in inputs and create a flat_fn for it. Dynamo currently cannot deal with pytree.tree_flatten and pytree.tree_unflatten, we therefore make it a HigherOrderOperator to trigger dynamo logic of handling HigherOrderOperators.
- "map_impl" is the actual operator that works with the rest of torch subsystems such as functionalization, make_fx. It accepts flattend arguments, and a num_mapped_args integer denoting how many of the flattend arguments need to mapped i.e. their first dimension will be unstacked.

2. We create the forward and backward graph in autograd key and call torch.autograd.Function. Currently, the backward graph is recomputation-based and we need to partition the joint graph in the future to be more efficient.

Example traced graphs for map operators:
### Case 1: simple f and autograd
```python
def f(x, y):
    return x + y

def g(xs, y):
    out = control_flow.map(f, xs, y)
    return torch.autograd.grad(out, (xs, y), torch.ones_like(out))

gm = make_fx(g, tracing_mode="symbolic")(torch.ones(3, 4, 5, requires_grad=True), torch.ones(5, requires_grad=True))
# gm.print_readable() produces following:
class g(torch.nn.Module):
    def forward(self, xs_1: f32[3, s1, s2], y_1: f32[s2]):
        # No stacktrace found for following nodes
        body_graph_0 = self.body_graph_0
        map_impl = torch.ops.map_impl(body_graph_0, 1, xs_1, y_1);  body_graph_0 = None
        getitem: f32[3, s1, s2] = map_impl[0];  map_impl = None
        ones_like: f32[3, s1, s2] = torch.ops.aten.ones_like.default(getitem, pin_memory = False)
        is_same_size = torch.ops.aten.is_same_size.default(getitem, ones_like);  getitem = None
        body_graph_1 = self.body_graph_1
        map_impl_1 = torch.ops.map_impl(body_graph_1, 2, xs_1, ones_like, y_1);  body_graph_1 = xs_1 = ones_like = None
        getitem_1 = map_impl_1[0]
        getitem_2: f32[3, s1, s2] = map_impl_1[1]
        getitem_3: f32[3, s2] = map_impl_1[2];  map_impl_1 = None
        sum_1: f32[1, s2] = torch.ops.aten.sum.dim_IntList(getitem_3, [0], True);  getitem_3 = None
        sym_size: Sym(s2) = torch.ops.aten.sym_size(y_1, 0);  y_1 = None
        view: f32[s2] = torch.ops.aten.view.default(sum_1, [sym_size]);  sum_1 = sym_size = None
        return (getitem_2, view)

    class <lambda>(torch.nn.Module):
        def forward(self, arg0_1, arg1_1: f32[s1, s2], arg2_1: f32[s2]):
            # No stacktrace found for following nodes
            add: f32[s1, s2] = torch.ops.aten.add.Tensor(arg1_1, arg2_1);  arg1_1 = arg2_1 = None
            return [add]

    class <lambda>(torch.nn.Module):
        def forward(self, arg0_1, arg1_1: f32[s1, s2], arg2_1: f32[s1, s2], arg3_1: f32[s2]):
            # No stacktrace found for following nodes
            add: f32[s1, s2] = torch.ops.aten.add.Tensor(arg1_1, arg3_1);  arg1_1 = None
            is_same_size = torch.ops.aten.is_same_size.default(add, arg2_1);  add = None
            sum_1: f32[1, s2] = torch.ops.aten.sum.dim_IntList(arg2_1, [0], True)
            sym_size: Sym(s2) = torch.ops.aten.sym_size(arg3_1, 0);  arg3_1 = None
            view: f32[s2] = torch.ops.aten.view.default(sum_1, [sym_size]);  sum_1 = sym_size = None
            return [None, arg2_1, view]
```
### Case 2: list input/output f and autograd
```python
def f(x, y):
    return [x[0].cos() + y.sin(), x[1].sin() * y.cos()]

def g(xs, y):
    out = control_flow.map(f, xs, y)
    flat_out, _ = pytree.tree_flatten(out)
    flat_inp, _ = pytree.tree_flatten((xs, y))
    requires_grad_inp = [inp for inp in flat_inp if inp.requires_grad]
    return torch.autograd.grad(flat_out, requires_grad_inp, [torch.ones_like(out) for out in flat_out])

gm = make_fx(g, tracing_mode="symbolic")(
    [torch.ones(3, 4, 5), torch.ones(3, 4, 5, requires_grad=True)],
    torch.ones(5, requires_grad=True))

# gm.print_readable() produces following:
class g(torch.nn.Module):
    def forward(self, xs, y):
        xs_1: f32[3, s1, s2], xs_2: f32[3, s1, s2], y_1: f32[s2], = fx_pytree.tree_flatten_spec([xs, y], self._in_spec)
        # No stacktrace found for following nodes
        body_graph_0 = self.body_graph_0
        map_impl = torch.ops.map_impl(body_graph_0, 2, xs_1, xs_2, y_1);  body_graph_0 = None
        getitem: f32[3, s1, s2] = map_impl[0]
        getitem_1: f32[3, s1, s2] = map_impl[1];  map_impl = None
        ones_like: f32[3, s1, s2] = torch.ops.aten.ones_like.default(getitem, pin_memory = False)
        ones_like_1: f32[3, s1, s2] = torch.ops.aten.ones_like.default(getitem_1, pin_memory = False)
        is_same_size = torch.ops.aten.is_same_size.default(getitem, ones_like);  getitem = None
        is_same_size_1 = torch.ops.aten.is_same_size.default(getitem_1, ones_like_1);  getitem_1 = None
        body_graph_1 = self.body_graph_1
        map_impl_1 = torch.ops.map_impl(body_graph_1, 4, xs_1, xs_2, ones_like, ones_like_1, y_1);  body_graph_1 = xs_1 = xs_2 = ones_like = ones_like_1 = None
        getitem_2 = map_impl_1[0]
        getitem_3 = map_impl_1[1]
        getitem_4: f32[3, s1, s2] = map_impl_1[2]
        getitem_5: f32[3, s2] = map_impl_1[3];  map_impl_1 = None
        sum_1: f32[1, s2] = torch.ops.aten.sum.dim_IntList(getitem_5, [0], True);  getitem_5 = None
        sym_size: Sym(s2) = torch.ops.aten.sym_size(y_1, 0);  y_1 = None
        view: f32[s2] = torch.ops.aten.view.default(sum_1, [sym_size]);  sum_1 = sym_size = None
        return pytree.tree_unflatten([getitem_4, view], self._out_spec)

    class <lambda>(torch.nn.Module):
        def forward(self, arg0_1, arg1_1: f32[s1, s2], arg2_1: f32[s1, s2], arg3_1: f32[s2]):
            # No stacktrace found for following nodes
            cos: f32[s1, s2] = torch.ops.aten.cos.default(arg1_1);  arg1_1 = None
            sin: f32[s2] = torch.ops.aten.sin.default(arg3_1)
            add: f32[s1, s2] = torch.ops.aten.add.Tensor(cos, sin);  cos = sin = None
            sin_1: f32[s1, s2] = torch.ops.aten.sin.default(arg2_1);  arg2_1 = None
            cos_1: f32[s2] = torch.ops.aten.cos.default(arg3_1);  arg3_1 = None
            mul: f32[s1, s2] = torch.ops.aten.mul.Tensor(sin_1, cos_1);  sin_1 = cos_1 = None
            return [add, mul]

    class <lambda>(torch.nn.Module):
        def forward(self, arg0_1, arg1_1: f32[s1, s2], arg2_1: f32[s1, s2], arg3_1: f32[s1, s2], arg4_1: f32[s1, s2], arg5_1: f32[s2]):
            # No stacktrace found for following nodes
            cos: f32[s1, s2] = torch.ops.aten.cos.default(arg1_1);  arg1_1 = None
            sin: f32[s2] = torch.ops.aten.sin.default(arg5_1)
            add: f32[s1, s2] = torch.ops.aten.add.Tensor(cos, sin);  cos = sin = None
            sin_1: f32[s1, s2] = torch.ops.aten.sin.default(arg2_1)
            cos_1: f32[s2] = torch.ops.aten.cos.default(arg5_1)
            mul: f32[s1, s2] = torch.ops.aten.mul.Tensor(sin_1, cos_1)
            is_same_size = torch.ops.aten.is_same_size.default(add, arg3_1);  add = None
            is_same_size_1 = torch.ops.aten.is_same_size.default(mul, arg4_1);  mul = None
            mul_1: f32[s1, s2] = torch.ops.aten.mul.Tensor(arg4_1, sin_1);  sin_1 = None
            mul_2: f32[s1, s2] = torch.ops.aten.mul.Tensor(arg4_1, cos_1);  arg4_1 = cos_1 = None
            sum_1: f32[1, s2] = torch.ops.aten.sum.dim_IntList(mul_1, [0], True);  mul_1 = None
            sym_size: Sym(s2) = torch.ops.aten.sym_size(arg5_1, 0)
            view: f32[s2] = torch.ops.aten.view.default(sum_1, [sym_size]);  sum_1 = None

            #
            sin_2: f32[s2] = torch.ops.aten.sin.default(arg5_1)
            neg: f32[s2] = torch.ops.aten.neg.default(sin_2);  sin_2 = None
            mul_3: f32[s2] = torch.ops.aten.mul.Tensor(view, neg);  view = neg = None
            cos_2: f32[s1, s2] = torch.ops.aten.cos.default(arg2_1);  arg2_1 = None
            mul_4: f32[s1, s2] = torch.ops.aten.mul.Tensor(mul_2, cos_2);  mul_2 = cos_2 = None
            sum_2: f32[1, s2] = torch.ops.aten.sum.dim_IntList(arg3_1, [0], True);  arg3_1 = None
            view_1: f32[s2] = torch.ops.aten.view.default(sum_2, [sym_size]);  sum_2 = sym_size = None
            cos_3: f32[s2] = torch.ops.aten.cos.default(arg5_1);  arg5_1 = None
            mul_5: f32[s2] = torch.ops.aten.mul.Tensor(view_1, cos_3);  view_1 = cos_3 = None
            add_1: f32[s2] = torch.ops.aten.add.Tensor(mul_3, mul_5);  mul_3 = mul_5 = None
            return [None, None, mul_4, add_1]
```

Pull Request resolved: https://github.com/pytorch/pytorch/pull/101633
Approved by: https://github.com/zou3519
 2023-05-17 16:52:26 +00:00
PyTorch MergeBot
e69198b043 Revert "Support map autograd and pytree in/out (#100494)" 
This reverts commit b8fa41be9d.

Reverted https://github.com/pytorch/pytorch/pull/100494 on behalf of https://github.com/PaliC due to breaking tests on trunk, please check hud.pytorch.org for the broken tests ([comment](https://github.com/pytorch/pytorch/pull/100494#issuecomment-1550454835))
 2023-05-16 22:50:18 +00:00
ydwu4
b8fa41be9d Support map autograd and pytree in/out (#100494) 
This PR adds autograd and pytree support for map operator.

Implementation-wise:

1. We temporarily make two HigherOrderOperators, "map" and "map_impl":
- "map" is user-facing. Currently, it unwraps the pytrees in inputs and create a flat_fn for it. Dynamo currently cannot deal with pytree.tree_flatten and pytree.tree_unflatten, we therefore make it a HigherOrderOperator to trigger dynamo logic of handling HigherOrderOperators.
- "map_impl" is the actual operator that works with the rest of torch subsystems such as functionalization, make_fx. It accepts flattend arguments, and a num_mapped_args integer denoting how many of the flattend arguments need to mapped i.e. their first dimension will be unstacked.

2. We create the forward and backward graph in autograd key and call torch.autograd.Function. Currently, the backward graph is recomputation-based and we need to partition the joint graph in the future to be more efficient.

Example traced graphs for map operators:
### Case 1: simple f and autograd
```python
def f(x, y):
    return x + y

def g(xs, y):
    out = control_flow.map(f, xs, y)
    return torch.autograd.grad(out, (xs, y), torch.ones_like(out))

gm = make_fx(g, tracing_mode="symbolic")(torch.ones(3, 4, 5, requires_grad=True), torch.ones(5, requires_grad=True))
# gm.print_readable() produces following:
class g(torch.nn.Module):
    def forward(self, xs_1: f32[3, s1, s2], y_1: f32[s2]):
        # No stacktrace found for following nodes
        body_graph_0 = self.body_graph_0
        map_impl = torch.ops.map_impl(body_graph_0, 1, xs_1, y_1);  body_graph_0 = None
        getitem: f32[3, s1, s2] = map_impl[0];  map_impl = None
        ones_like: f32[3, s1, s2] = torch.ops.aten.ones_like.default(getitem, pin_memory = False)
        is_same_size = torch.ops.aten.is_same_size.default(getitem, ones_like);  getitem = None
        body_graph_1 = self.body_graph_1
        map_impl_1 = torch.ops.map_impl(body_graph_1, 2, xs_1, ones_like, y_1);  body_graph_1 = xs_1 = ones_like = None
        getitem_1 = map_impl_1[0]
        getitem_2: f32[3, s1, s2] = map_impl_1[1]
        getitem_3: f32[3, s2] = map_impl_1[2];  map_impl_1 = None
        sum_1: f32[1, s2] = torch.ops.aten.sum.dim_IntList(getitem_3, [0], True);  getitem_3 = None
        sym_size: Sym(s2) = torch.ops.aten.sym_size(y_1, 0);  y_1 = None
        view: f32[s2] = torch.ops.aten.view.default(sum_1, [sym_size]);  sum_1 = sym_size = None
        return (getitem_2, view)

    class <lambda>(torch.nn.Module):
        def forward(self, arg0_1, arg1_1: f32[s1, s2], arg2_1: f32[s2]):
            # No stacktrace found for following nodes
            add: f32[s1, s2] = torch.ops.aten.add.Tensor(arg1_1, arg2_1);  arg1_1 = arg2_1 = None
            return [add]

    class <lambda>(torch.nn.Module):
        def forward(self, arg0_1, arg1_1: f32[s1, s2], arg2_1: f32[s1, s2], arg3_1: f32[s2]):
            # No stacktrace found for following nodes
            add: f32[s1, s2] = torch.ops.aten.add.Tensor(arg1_1, arg3_1);  arg1_1 = None
            is_same_size = torch.ops.aten.is_same_size.default(add, arg2_1);  add = None
            sum_1: f32[1, s2] = torch.ops.aten.sum.dim_IntList(arg2_1, [0], True)
            sym_size: Sym(s2) = torch.ops.aten.sym_size(arg3_1, 0);  arg3_1 = None
            view: f32[s2] = torch.ops.aten.view.default(sum_1, [sym_size]);  sum_1 = sym_size = None
            return [None, arg2_1, view]
```
### Case 2: list input/output f and autograd
```python
def f(x, y):
    return [x[0].cos() + y.sin(), x[1].sin() * y.cos()]

def g(xs, y):
    out = control_flow.map(f, xs, y)
    flat_out, _ = pytree.tree_flatten(out)
    flat_inp, _ = pytree.tree_flatten((xs, y))
    requires_grad_inp = [inp for inp in flat_inp if inp.requires_grad]
    return torch.autograd.grad(flat_out, requires_grad_inp, [torch.ones_like(out) for out in flat_out])

gm = make_fx(g, tracing_mode="symbolic")(
    [torch.ones(3, 4, 5), torch.ones(3, 4, 5, requires_grad=True)],
    torch.ones(5, requires_grad=True))

# gm.print_readable() produces following:
class g(torch.nn.Module):
    def forward(self, xs, y):
        xs_1: f32[3, s1, s2], xs_2: f32[3, s1, s2], y_1: f32[s2], = fx_pytree.tree_flatten_spec([xs, y], self._in_spec)
        # No stacktrace found for following nodes
        body_graph_0 = self.body_graph_0
        map_impl = torch.ops.map_impl(body_graph_0, 2, xs_1, xs_2, y_1);  body_graph_0 = None
        getitem: f32[3, s1, s2] = map_impl[0]
        getitem_1: f32[3, s1, s2] = map_impl[1];  map_impl = None
        ones_like: f32[3, s1, s2] = torch.ops.aten.ones_like.default(getitem, pin_memory = False)
        ones_like_1: f32[3, s1, s2] = torch.ops.aten.ones_like.default(getitem_1, pin_memory = False)
        is_same_size = torch.ops.aten.is_same_size.default(getitem, ones_like);  getitem = None
        is_same_size_1 = torch.ops.aten.is_same_size.default(getitem_1, ones_like_1);  getitem_1 = None
        body_graph_1 = self.body_graph_1
        map_impl_1 = torch.ops.map_impl(body_graph_1, 4, xs_1, xs_2, ones_like, ones_like_1, y_1);  body_graph_1 = xs_1 = xs_2 = ones_like = ones_like_1 = None
        getitem_2 = map_impl_1[0]
        getitem_3 = map_impl_1[1]
        getitem_4: f32[3, s1, s2] = map_impl_1[2]
        getitem_5: f32[3, s2] = map_impl_1[3];  map_impl_1 = None
        sum_1: f32[1, s2] = torch.ops.aten.sum.dim_IntList(getitem_5, [0], True);  getitem_5 = None
        sym_size: Sym(s2) = torch.ops.aten.sym_size(y_1, 0);  y_1 = None
        view: f32[s2] = torch.ops.aten.view.default(sum_1, [sym_size]);  sum_1 = sym_size = None
        return pytree.tree_unflatten([getitem_4, view], self._out_spec)

    class <lambda>(torch.nn.Module):
        def forward(self, arg0_1, arg1_1: f32[s1, s2], arg2_1: f32[s1, s2], arg3_1: f32[s2]):
            # No stacktrace found for following nodes
            cos: f32[s1, s2] = torch.ops.aten.cos.default(arg1_1);  arg1_1 = None
            sin: f32[s2] = torch.ops.aten.sin.default(arg3_1)
            add: f32[s1, s2] = torch.ops.aten.add.Tensor(cos, sin);  cos = sin = None
            sin_1: f32[s1, s2] = torch.ops.aten.sin.default(arg2_1);  arg2_1 = None
            cos_1: f32[s2] = torch.ops.aten.cos.default(arg3_1);  arg3_1 = None
            mul: f32[s1, s2] = torch.ops.aten.mul.Tensor(sin_1, cos_1);  sin_1 = cos_1 = None
            return [add, mul]

    class <lambda>(torch.nn.Module):
        def forward(self, arg0_1, arg1_1: f32[s1, s2], arg2_1: f32[s1, s2], arg3_1: f32[s1, s2], arg4_1: f32[s1, s2], arg5_1: f32[s2]):
            # No stacktrace found for following nodes
            cos: f32[s1, s2] = torch.ops.aten.cos.default(arg1_1);  arg1_1 = None
            sin: f32[s2] = torch.ops.aten.sin.default(arg5_1)
            add: f32[s1, s2] = torch.ops.aten.add.Tensor(cos, sin);  cos = sin = None
            sin_1: f32[s1, s2] = torch.ops.aten.sin.default(arg2_1)
            cos_1: f32[s2] = torch.ops.aten.cos.default(arg5_1)
            mul: f32[s1, s2] = torch.ops.aten.mul.Tensor(sin_1, cos_1)
            is_same_size = torch.ops.aten.is_same_size.default(add, arg3_1);  add = None
            is_same_size_1 = torch.ops.aten.is_same_size.default(mul, arg4_1);  mul = None
            mul_1: f32[s1, s2] = torch.ops.aten.mul.Tensor(arg4_1, sin_1);  sin_1 = None
            mul_2: f32[s1, s2] = torch.ops.aten.mul.Tensor(arg4_1, cos_1);  arg4_1 = cos_1 = None
            sum_1: f32[1, s2] = torch.ops.aten.sum.dim_IntList(mul_1, [0], True);  mul_1 = None
            sym_size: Sym(s2) = torch.ops.aten.sym_size(arg5_1, 0)
            view: f32[s2] = torch.ops.aten.view.default(sum_1, [sym_size]);  sum_1 = None

            #
            sin_2: f32[s2] = torch.ops.aten.sin.default(arg5_1)
            neg: f32[s2] = torch.ops.aten.neg.default(sin_2);  sin_2 = None
            mul_3: f32[s2] = torch.ops.aten.mul.Tensor(view, neg);  view = neg = None
            cos_2: f32[s1, s2] = torch.ops.aten.cos.default(arg2_1);  arg2_1 = None
            mul_4: f32[s1, s2] = torch.ops.aten.mul.Tensor(mul_2, cos_2);  mul_2 = cos_2 = None
            sum_2: f32[1, s2] = torch.ops.aten.sum.dim_IntList(arg3_1, [0], True);  arg3_1 = None
            view_1: f32[s2] = torch.ops.aten.view.default(sum_2, [sym_size]);  sum_2 = sym_size = None
            cos_3: f32[s2] = torch.ops.aten.cos.default(arg5_1);  arg5_1 = None
            mul_5: f32[s2] = torch.ops.aten.mul.Tensor(view_1, cos_3);  view_1 = cos_3 = None
            add_1: f32[s2] = torch.ops.aten.add.Tensor(mul_3, mul_5);  mul_3 = mul_5 = None
            return [None, None, mul_4, add_1]
```

Pull Request resolved: https://github.com/pytorch/pytorch/pull/100494
Approved by: https://github.com/zou3519
 2023-05-16 22:05:11 +00:00
Nikita Karetnikov
42e65a2587 [pt2] add meta for linalg_lu_factor_ex (#101375) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/101375
Approved by: https://github.com/lezcano
 2023-05-16 20:56:54 +00:00
Khushi
01c7106580 [opinfo] empty_strided (#100890) 
Follows: #100223

Pull Request resolved: https://github.com/pytorch/pytorch/pull/100890
Approved by: https://github.com/ezyang
 2023-05-15 23:39:39 +00:00
Nikita Karetnikov
9eb1748b2b [pt2] add meta and SymInt support for linalg_lu (#101372) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/101372
Approved by: https://github.com/lezcano, https://github.com/albanD
 2023-05-15 20:25:00 +00:00
Nikita Karetnikov
ac4cc63ae2 [pt2] add meta for linalg_ldl_solve (#101367) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/101367
Approved by: https://github.com/lezcano
 2023-05-15 20:25:00 +00:00
Nikita Karetnikov
7dd8e08817 [pt2] add meta for linalg_ldl_factor_ex (#101362) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/101362
Approved by: https://github.com/lezcano
 2023-05-15 02:56:49 +00:00
Nikita Karetnikov
a8964d6377 [pt2] add meta and SymInt support for linalg_householder_product (#101315) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/101315
Approved by: https://github.com/lezcano
 2023-05-15 02:56:49 +00:00
Nikita Karetnikov
6abde61f8e [pt2] add meta function for _linalg_eigh (#100964) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100964
Approved by: https://github.com/ezyang
 2023-05-10 15:45:15 +00:00
Khushi
51fe53e619 [opinfo] item (#100313) 
Follows #100223

Pull Request resolved: https://github.com/pytorch/pytorch/pull/100313
Approved by: https://github.com/ezyang
 2023-05-10 11:32:45 +00:00
Nikita Karetnikov
1e591a8b64 [pt2] add meta function for solve_triangular (#100829) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100829
Approved by: https://github.com/ezyang
 2023-05-08 13:48:15 +00:00
Nikita Karetnikov
266c84e3ab [pt2] add meta function for linalg_qr (#100714) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100714
Approved by: https://github.com/ezyang, https://github.com/lezcano
 2023-05-06 15:04:02 +00:00
Nikita Karetnikov
37f1be041a [pt2] enable svd in fake_tensor (#100130) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100130
Approved by: https://github.com/ezyang, https://github.com/lezcano
 2023-05-05 06:27:59 +00:00
Michael Voznesensky
fe3ecfe0cf Add AotAutogradFallbackTests to dynamic suite (#100454) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100454
Approved by: https://github.com/ezyang
 2023-05-04 04:28:45 +00:00
Nikita Karetnikov
e87ed2a88d [primTorch] add ref for polar (#100345) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100345
Approved by: https://github.com/ezyang
 2023-05-04 01:37:02 +00:00
Nikita Karetnikov
279f3cd0a6 [pt2] add SymInt support for dsplit, hsplit, vsplit (#100352) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100352
Approved by: https://github.com/Skylion007, https://github.com/ezyang
 2023-05-02 18:51:03 +00:00
Nikita Karetnikov
41361538a9 [pt2] add SymInt support for tensordot and inner (#100356) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100356
Approved by: https://github.com/ezyang
 2023-05-02 14:42:50 +00:00
Brian Hirsh
62fad315c1 fix per-dispatchkey-mode caching bug (#98030) 
The bug was that: if you want to move a mode to the autograd key, we need to use the "functionality" key for it (AutogradFunctionality). But when we do that, we need to clear any PythonDispatcher caches for every op for **every** autograd key (since you could run autograd ops with both cpu and cuda tensors underneath the mode, which both may have been cached).

I didn't add a test, since this ends up getting indirectly tests by export in the PR. If someone would prefer a direct test I can add one.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/98030
Approved by: https://github.com/ezyang
 2023-04-25 21:58:14 +00:00
Aaron Gokaslan
e2a3817dfd [BE] Enable C419 rule for any all shortcircuiting (#99890) 
Apparently https://github.com/pytorch/pytorch/pull/78142 made torch.JIT allow for simple generator expressions which allows us to enable rules that replace unnecessary list comprehensions with generators in any/all. This was originally part of #99280 but I split it off into this PR so that it can be easily reverted should anything break.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/99890
Approved by: https://github.com/justinchuby, https://github.com/kit1980, https://github.com/malfet
 2023-04-25 15:02:13 +00:00
Nikita Karetnikov
fbb0ff10a4 [pt2] add SymInt support for trapezoid ops (#99281) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/99281
Approved by: https://github.com/ezyang
 2023-04-25 00:44:25 +00:00
Wanchao Liang
ff7d5b62d4 Improve ProxyTensor tensor_tree list/tuple handling (#99897) 
This PR improves the list/tuple handling by merging the logic into
`wrap_with_proxy` directly, and set_meta when we find the current
proxy is a fx.Proxy. This also solves the problem that even `fused_adam`
have `val`, some corresponding `getitem` calls followed after `fused_adam` don't have val
Pull Request resolved: https://github.com/pytorch/pytorch/pull/99897
Approved by: https://github.com/ezyang
 2023-04-24 22:50:02 +00:00
Michael Voznesensky
4c2892944f Guard static shapes alongside tensors, instead of from shape_env, in dynamic_shapes=True (#99566) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/99566
Approved by: https://github.com/ezyang
 2023-04-22 16:46:52 +00:00
Edward Z. Yang
10c938abef Handle meta['val'] for tuple of lists. (#99724) 
Fixes https://github.com/pytorch/pytorch/issues/99356

Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/99724
Approved by: https://github.com/wanchaol
 2023-04-21 22:33:21 +00:00
Elias Ellison
638feec4e3 Turn on meta converter for complex (#98869) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98869
Approved by: https://github.com/ngimel
 2023-04-20 16:42:38 +00:00
Richard Zou
44b09bf673 Reland "Simple Custom Operator API, V0 (#98440)" (#99416) 
See the original PR (#98440) for the description. It broke internal
builds due to proxy_tensor.py not importing torch._dynamo, which is
being fixed in the previous PR in the stack.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/99416
Approved by: https://github.com/soulitzer, https://github.com/bdhirsh
 2023-04-18 23:48:33 +00:00
PyTorch MergeBot
f497031df9 Revert "Simple Custom Operator API, V0 (#98440)" 
This reverts commit 0157b2d722.

Reverted https://github.com/pytorch/pytorch/pull/98440 on behalf of https://github.com/DanilBaibak due to Break internal build
 2023-04-18 13:04:27 +00:00
Nikita Karetnikov
106ccf4a2a [pt2] add meta function for linalg.cross (#99279) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/99279
Approved by: https://github.com/ezyang
 2023-04-17 21:21:45 +00:00
Nikita Karetnikov
6f7b434f7b [pt2] add SymInt support for column_stack (#99276) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/99276
Approved by: https://github.com/ezyang
 2023-04-17 21:21:45 +00:00
PyTorch MergeBot
08dd4ad0b9 Revert "[pt2] add SymInt support for column_stack (#99276)" 
This reverts commit 775dd869d0.

Reverted https://github.com/pytorch/pytorch/pull/99276 on behalf of https://github.com/ezyang due to reverting this one too for safety
 2023-04-17 19:37:58 +00:00
PyTorch MergeBot
f957334c2b Revert "[pt2] add meta function for linalg.cross (#99279)" 
This reverts commit efc3887ea5.

Reverted https://github.com/pytorch/pytorch/pull/99279 on behalf of https://github.com/ezyang due to Apparently this is breaking inductor on master? So weird
 2023-04-17 19:33:16 +00:00
Tugsbayasgalan Manlaibaatar
7401f0f8ce Add unbacked symbool support (#98877) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98877
Approved by: https://github.com/ezyang
 2023-04-17 17:45:10 +00:00
Richard Zou
0157b2d722 Simple Custom Operator API, V0 (#98440) 
This PR introduces CustomOp, a wrapper around a dispatcher operator that allows
users to define custom operators. It adds the skeleton for CustomOp and
some very simple behavior: as of this PR:
- one can create a CustomOp for an operator that does not have inplace or aliasing
- give it CPU/CUDA and Meta implementations
- and trace it into a graph via make_fx.

The design follows
https://docs.google.com/document/d/19Uc5OUCA187q9BZggJb70RT2ZoSTDoG5QQkJkZwd25M/edit
Concretely, we implement the following things mentioned in the doc in this PR:
- Entrypoint 1 (CustomOp.define, creating a new custom operator)
- impl (to define device-specific code) and impl_meta (to define meta
formulas)

The goal for the short term is to get the code to a state where it can be trialed
by the export folks. On top of this PR, the blockers are:
- adding Entrypoint 3 (CustomOp.from_existing)
- adding a way to do data-dependent shape formulas
These will come in future PRs since this one is getting long.

Things that will come in the longer-near-term (before 2.1):
- adding the other entrypoints mentioned in the doc (2 & 3)
- more safety checks and better error messages
- support for views and mutation
- support for defining autograd formulas
- support for functionalization
- making this API public (it's private right now).

Test Plan:
- added a new test case, TestCustomOp. It mostly tests a bunch of error
cases.
- added OpInfos for custom operators and hooked these up to
test_proxy_tensor to test that they work with make_fx. These custom
operators were based off of the ones in the autograd_function_db.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98440
Approved by: https://github.com/ezyang
 2023-04-17 12:17:32 +00:00
Nikita Karetnikov
efc3887ea5 [pt2] add meta function for linalg.cross (#99279) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/99279
Approved by: https://github.com/ezyang
 2023-04-17 03:05:20 +00:00
Nikita Karetnikov
775dd869d0 [pt2] add SymInt support for column_stack (#99276) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/99276
Approved by: https://github.com/ezyang
 2023-04-17 03:05:20 +00:00
Nikita Karetnikov
21681f36f4 [pt2] add SymInt support for fft ops (#99115) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/99115
Approved by: https://github.com/ezyang
 2023-04-15 18:01:39 +00:00
Peter Bell
7b91bd2a7b [primTorch] Add count_nonzero (#98995) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98995
Approved by: https://github.com/lezcano
 2023-04-13 22:08:19 +00:00
Nikita Karetnikov
8db04e080c [pt2] add SymInt support for cdist (#98881) 
Fixes #98853.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98881
Approved by: https://github.com/ezyang
 2023-04-12 23:06:40 +00:00
Edward Z. Yang
419ad49e65 Make Tensor.__contains__ accept SymInt/Float/Bool. (#98933) 
Fixes https://github.com/pytorch/pytorch/issues/98870

Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98933
Approved by: https://github.com/albanD, https://github.com/Skylion007
 2023-04-12 19:16:33 +00:00
Nikita Karetnikov
ff825de442 [primTorch] add ref for cumprod (#98670) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98670
Approved by: https://github.com/ezyang
 2023-04-09 15:22:28 +00:00
Nikita Karetnikov
a2e7910dfd [pt2] remove skip for masked.logsumexp in test_proxy_tensor.py (#98676) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98676
Approved by: https://github.com/ezyang
 2023-04-09 01:28:16 +00:00
Nikita Karetnikov
b411238d76 [pt2] add meta function for logcumsumexp (#98683) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98683
Approved by: https://github.com/ezyang
 2023-04-09 01:26:37 +00:00
Nikita Karetnikov
1c226f5aad [pt2] add meta functions for cummax and cummin (#98552) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98552
Approved by: https://github.com/Chillee
 2023-04-07 17:58:28 +00:00
Nikita Karetnikov
7b25976323 [pt2] add meta function for take (#98451) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98451
Approved by: https://github.com/ezyang
 2023-04-06 14:48:35 +00:00
Michael Voznesensky
b1e60bfb6a Pass f_locals as a dict rather than kwargs (#98107) 
Fixes https://github.com/pytorch/pytorch/issues/97688

One big problem is that instead of printing x < y we now print
`E["x"] < E["y"]` and now all of the tests wobbled and I'm mad.

Signed-off-by: Edward Z. Yang <ezyangmeta.com>

Pull Request resolved: https://github.com/pytorch/pytorch/pull/98107
Approved by: https://github.com/ezyang
 2023-04-04 00:30:08 +00:00
Edward Z. Yang
8372c5dc68 Refactor dynamic dims api, stateless internals, higher level export API (#96699) 
The purpose of this API is to execute a few large components of work:

1) Refactor all the internals of plumbing dynamic dimension information after dynamo to be stateless
2) Decouple allocation controls around dynamic dimensions from verification
3) For (2), for allocation, create an enum that dictates whether we are in DUCK (default today), STATIC (aka assume_static_default in the past), or DYNAMIC (aka user constrained, do not duck shape)
4) For (2), for verification, we separate out the list of dynamic ranges entirely from allocation. This means shape_env does not tracking for what we verify on, and instead, it is the callers job to invoke produce_guards() with the various things they want verified, specifically, with the valid ranges. We do use constrain ranges to refine value ranges when doing analysis.
5) We have decided, therefore, as an extension of (4) to double down on "late" checks versus "eager" checks, primarily because the mechanisms for gathering what actually matters happens during guards, and should be a purview of the caller seeking guards, not the shape env. However, for dynamo, these structures are essentially one and the same.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/96699
Approved by: https://github.com/avikchaudhuri, https://github.com/ezyang
 2023-03-29 16:55:49 +00:00
Brian Hirsh
35c9ea89fa dont bake in defaults when tracing *_like factories (#97564) 
quick fix for https://github.com/pytorch/pytorch/issues/97541. letting CI run to see if there's any fallout

Pull Request resolved: https://github.com/pytorch/pytorch/pull/97564
Approved by: https://github.com/ezyang
 2023-03-27 22:53:44 +00:00
Brian Hirsh
af440c427b [draft for discussion] add per-dispatch key modes (#97052) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/97052
Approved by: https://github.com/ezyang, https://github.com/zou3519
 2023-03-21 23:45:45 +00:00
Rohan Gupta
b01d6f2cdb addmv decomp #2 (#96264) 
Fixes #94617

Pull Request resolved: https://github.com/pytorch/pytorch/pull/96264
Approved by: https://github.com/ngimel, https://github.com/ezyang
 2023-03-16 23:09:45 +00:00
Nikita Karetnikov
0d7c44096a Add baddbmm meta function (#96548) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/96548
Approved by: https://github.com/ezyang
 2023-03-11 19:09:24 +00:00
Nikita Karetnikov
8e0d5bf538 [primTorch] add meta implementation for aten.min.dim (#96442) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/96442
Approved by: https://github.com/ngimel
 2023-03-11 18:51:51 +00:00
Edward Z. Yang
98ff841a75 Use maxint to bound integers. (#96121) 
We don't actually support arbitrary precision integers.

Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/96121
Approved by: https://github.com/tugsbayasgalan, https://github.com/lezcano
 2023-03-07 12:46:19 +00:00
Edward Z. Yang
680214ac11 SymIntify a few more relatively non-controversial schemas (#96100) 
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/96100
Approved by: https://github.com/Skylion007
 2023-03-06 23:12:40 +00:00
Jason Ansel
5dd52e250f [inductor] Add some simple decomps (#96039) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/96039
Approved by: https://github.com/ngimel
 2023-03-05 17:07:56 +00:00
Edward Z. Yang
027ebca4d7 Don't use guardless contiguity/stride-like implementations (#95733) 
These prevent us from simplifying tests involving unbacked SymInts,
and then you end up with unbacked SymInt in guards, which is bad.

Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/95733
Approved by: https://github.com/tugsbayasgalan
 2023-03-03 21:56:41 +00:00
PyTorch MergeBot
4026c62174 Revert "Don't use guardless contiguity/stride-like implementations (#95733)" 
This reverts commit deaf077de8.

Reverted https://github.com/pytorch/pytorch/pull/95733 on behalf of https://github.com/ezyang due to apparently this regresses executorch tests internally
 2023-03-03 17:43:05 +00:00
Edward Z. Yang
deaf077de8 Don't use guardless contiguity/stride-like implementations (#95733) 
These prevent us from simplifying tests involving unbacked SymInts,
and then you end up with unbacked SymInt in guards, which is bad.

Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/95733
Approved by: https://github.com/tugsbayasgalan
 2023-03-01 23:14:58 +00:00
Edward Z. Yang
e628a3e724 Don't generate guards that refer to unbacked SymInts (#95732) 
This regresses unbacked batch resnet, but I have a plan to recover that
too.

Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/95732
Approved by: https://github.com/tugsbayasgalan
 2023-03-01 06:14:27 +00:00
Edward Z. Yang
d78274b759 Automatically guard when SymInt is converted to int (#95479) 
During enablement, we disabled int() conversions because they were
any easy way to footgun guards.  We have enough of dynamic shapes
working now that this is now causing spurious errors; e.g., if you feed
a symbolic int to x.size(symint).  We now allow for implicit conversions
of SymInt to int here, posting a guard.  We expect guard provenance
to help people debug overspecialization.

Fixes https://github.com/pytorch/pytorch/issues/95328

Signed-off-by: Edward Z. Yang <ezyang@meta.com>

Pull Request resolved: https://github.com/pytorch/pytorch/pull/95479
Approved by: https://github.com/wconstab, https://github.com/voznesenskym, https://github.com/ngimel
 2023-02-25 19:41:51 +00:00
Edward Z. Yang
8efe4fd590 Memoize repeated nonzero calls to the same fake tensor (#95399) 
This removes the need to explicitly constrain_unify `x[mask]` and `y[mask]` when mask is a boolean tensor. It's very narrow but it seems to work in practice.

To invalidate the nonzero call when mutation occurs, I use version counter. I know there are ways to bypass this but I think it's good enough for now.

Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/95399
Approved by: https://github.com/eellison
 2023-02-24 00:27:45 +00:00
Edward Z. Yang
4833e47feb Add support for nonzero, some improvements to reduce guards (#95387) 
This takes the strategy described in https://docs.google.com/document/d/1lFRYAJo5nrfxRhwIzGnfi2pbLpU6T4ytSRSuLJ5qebI/edit#

It is essentially https://github.com/pytorch/pytorch/pull/95222 but squashed and with changes that are unnecessary given that we assume nonzero returns > 1.

What's in the PR:

* nonzero now supports meta propagation. When `capture_dynamic_output_shape_ops`, it will return a tensor with an unbacked SymInt representing the size in question.
* The unbacked SymInt is UNSOUNDLY assumed to be not equal to 0/1. We will still error if you guard otherwise.
* PrimTorch pointwise operators are updated to use empty_permuted, to avoid guarding on unbacked SymInt from empty_strided (tested in `test_dynamic_pointwise_scalar`)
* Convolution is updated to skip backend selection if batch is unbacked, to avoid guarding on unbacked SymInt (tested in `test_unbacked_batch_resnet`)
* I kept the helper utilities like `definitely_true` for working with possibly unbacked SymInts. They're not used right now but maybe someone will find them useful.
* Added `constrain_unify` to let you specify two unbacked SymInts must have the same value

Signed-off-by: Edward Z. Yang <ezyang@meta.com>

Pull Request resolved: https://github.com/pytorch/pytorch/pull/95387
Approved by: https://github.com/voznesenskym
 2023-02-24 00:27:45 +00:00
Edward Z. Yang
3758559a58 Reland "Introduce constrain_range; remove old expr_subs (#95063)" (#95209) 
This reverts commit 4e88547c95.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/95209
Approved by: https://github.com/albanD
 2023-02-22 18:16:25 +00:00
PyTorch MergeBot
cf6e078c34 Revert "Reland "Introduce constrain_range; remove old expr_subs (#95063)" (#95209)" 
This reverts commit f7bf31fff1.

Reverted https://github.com/pytorch/pytorch/pull/95209 on behalf of https://github.com/ezyang due to internal sympy is too old
 2023-02-22 01:58:58 +00:00
Edward Z. Yang
f7bf31fff1 Reland "Introduce constrain_range; remove old expr_subs (#95063)" (#95209) 
This reverts commit 4e88547c95.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/95209
Approved by: https://github.com/albanD
 2023-02-21 18:02:48 +00:00