[export] handle new roots & root swapping in derived dims suggested fixes #125543

pianpwk · 2024-05-05T00:37:28Z

Pull Request resolved: #125543

This PR address 2 issues with derived dim suggested fixes, 1) newly introduced roots, and 2) root swapping.

1 | Newly introduced roots appear with modulo guards, e.g. Mod(dx, 2) = 0 suggests dx is a derived dim equal to 2 * _dx, introducing a new root _dx. Currently the final suggested fixes handle this correctly, but we can get intermediate results where related derived dims don't rely on a unified root, and are a mixture of min/max range and derived suggestions.

For example:

"dx": {"eq": 3*_dx-1, "max": 36}
"dy": {"eq": dx+1}
This should lead to suggested fixes
  _dx = Dim('_dx', max=12)
  dx = 3 * _dx - 1
  dy = 3 * _dx

This PR prettifies the suggested fixes routine by unifying to a single root, and making each intermediate suggestion either a derived dim or min/max range, not both.

2 | The current suggested fixes for derived dims can lead to root dims/derived dims being swapped, e.g. dy - 1, dy -> dx, dx + 1. This leads to problematic suggested fixes that look like dy - 1 = Dim("dy - 1") since we don't have access to the original variable name.

This PR only adds a suggested fix for the root dim, and removes all other derived suggestions.

For example, with the export test case test_derived_dim_out_of_order_simplified:

_dimz = torch.export.Dim("_dimz", min=6, max=8)
dimy = _dimz - 1
dimx = dimy - 1
dimz = torch.export.Dim("dimz", min=6, max=8)  # doesn't work, should be = _dimz

class Foo(torch.nn.Module):
    def forward(self, x, y, z):
        return x + y[1:] + z[2:]

foo = Foo()
u, v, w = torch.randn(5), torch.randn(6), torch.randn(7)
export(
    foo,
    (u, v, w),
    dynamic_shapes=({0: dimx}, {0: dimy}, {0: dimz}),
)

Before:

Suggested fixes:
  _dimz = Dim('_dimz', min=3, max=9223372036854775807)  # 2 <= _dimz - 1 <= 9223372036854775806
  _dimz - 2 = Dim('_dimz - 2', min=4, max=6)
  _dimz = Dim('_dimz', min=2, max=9223372036854775806)  # 2 <= _dimz <= 9223372036854775806
  _dimz - 1 = _dimz - 1
  dimz = _dimz

New suggested fixes:

Suggested fixes:
  dimz = _dimz

Note: This assumes the specified derived relations between dims are correct. This should be valid because: 1) if the relation is plain wrong (e.g. (dx, dx - 1) provided with inputs (6, 4)), this gets caught in beforehand in produce_guards. 2) if the relation is correct but does not match the emitted guard, for example:

def forward(self, x, y):
    return x.reshape([-1]) + y  # guard: s0 * 2 = s1
dx = Dim("dx")
export(
    model,
    (torch.randn(6, 2), torch.randn(12)),
    dynamic_shapes={"x": (dx, 2), "y": (dx + 6, )}
)

This produces two linear equations, leading to specialization since a) produce_guards is able to solve for a concrete value, and b) the export constraint solver will anyways force specializations due to range constraints.
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @chauhang

pytorch-bot · 2024-05-05T00:37:30Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/125543

📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

Note: Links to docs will display an error until the docs builds have been completed.

✅ You can merge normally! (1 Unrelated Failure)

As of commit aff51b8 with merge base cbb79a2 ():

UNSTABLE - The following job failed but was likely due to flakiness present on trunk and has been marked as unstable:

inductor / linux-jammy-cpu-py3.8-gcc11-inductor / test (inductor_torchbench_cpu_smoketest_perf, 1, 1, linux.24xl.spr-metal, unstable) (gh) (#126993)
Process completed with exit code 1.

This comment was automatically generated by Dr. CI and updates every 15 minutes.

facebook-github-bot · 2024-05-05T00:37:40Z

This pull request was exported from Phabricator. Differential Revision: D56974543

facebook-github-bot · 2024-05-05T05:35:53Z

This pull request was exported from Phabricator. Differential Revision: D56974543

Summary: The current suggested fixes for derived dims can lead to root dims/derived dims being swapped, e.g. `dy - 1, dy` -> `dx, dx + 1`. This leads to problematic suggested fixes that look like `dy - 1 = Dim("dy - 1")` since we don't have access to the original variable name. This PR only adds a suggested fix for the root dim, and removes all other derived suggestions. For example, with the export test case test_derived_dim_out_of_order_simplified: ``` _dimz = torch.export.Dim("_dimz", min=6, max=8) dimy = _dimz - 1 dimx = dimy - 1 dimz = torch.export.Dim("dimz", min=6, max=8) # doesn't work, should be = _dimz class Foo(torch.nn.Module): def forward(self, x, y, z): return x + y[1:] + z[2:] foo = Foo() u, v, w = torch.randn(5), torch.randn(6), torch.randn(7) export( foo, (u, v, w), dynamic_shapes=({0: dimx}, {0: dimy}, {0: dimz}), ) ``` Before: ``` Suggested fixes: _dimz = Dim('_dimz', min=3, max=9223372036854775807) # 2 <= _dimz - 1 <= 9223372036854775806 _dimz - 2 = Dim('_dimz - 2', min=4, max=6) _dimz = Dim('_dimz', min=2, max=9223372036854775806) # 2 <= _dimz <= 9223372036854775806 _dimz - 1 = _dimz - 1 dimz = _dimz ``` New suggested fixes: ``` Suggested fixes: dimz = _dimz ``` Note: This assumes the specified derived relations between dims are correct. This should be valid because: 1) if the relation is plain wrong (e.g. (dx, dx - 1) provided with inputs (6, 4)), this gets caught in beforehand in produce_guards. 2) if the relation is correct but does not match the emitted guard, for example: ``` def forward(self, x, y): return x.reshape([-1]) + y # guard: s0 * 2 = s1 dx = Dim("dx") export( model, (torch.randn(6, 2), torch.randn(12)), dynamic_shapes={"x": (dx, 2), "y": (dx + 6, )} ) ``` This produces two linear equations, leading to specialization since a) produce_guards is able to solve for a concrete value, and b) the export constraint solver will anyways force specializations due to range constraints. Test Plan: export test cases Differential Revision: D56974543

facebook-github-bot · 2024-05-06T16:05:00Z

This pull request was exported from Phabricator. Differential Revision: D56974543

avikchaudhuri · 2024-05-06T16:36:52Z

torch/fx/experimental/symbolic_shapes.py

    int_sym = _sympy_cast_symbool_to_symint_guardless(symbool.node.expr)
    return symbool.node.shape_env.create_symintnode(int_sym, hint=int(symbool.node.require_hint()) if has_hint(symbool) else None)

+# use class name to avoid circular imports


eh, couldn't you do a local import to avoid circular imports?

avikchaudhuri · 2024-05-06T16:37:36Z

torch/fx/experimental/symbolic_shapes.py

+def _is_dim(dim):
+    return dim.__class__.__name__ == "_Dim"
+
+def _solve_for_root(expr, val):


could you use try_solve instead?

avikchaudhuri · 2024-05-06T16:43:06Z

torch/fx/experimental/symbolic_shapes.py

+            if k in name_to_dim and _is_derived_dim(name_to_dim[k])
+        }
+
+        # evaluate each root for either concrete val or min/max range


The algo here is hard to follow, please add a running example. It's intricate logic, having the transformation explicitly laid out in comments, and how it's achieved with every major step of the transformation, would be good.

Also maybe add comments on why just suggesting fixes to the roots is sufficient.

avikchaudhuri · 2024-05-06T16:45:37Z

torch/fx/experimental/symbolic_shapes.py


+            # retrieve dynamic shapes
+            name_to_dim = {}
+            for dim in pytree.tree_flatten(


Isn't there a different way to collect all dims? Like, have you checked the contents of source_name_to_debug_name?

I think the main issue here is collecting root dims which aren't attached to a size

…ixes (#125543) Summary: This PR address 2 issues with derived dim suggested fixes, 1) newly introduced roots, and 2) root swapping. 1) Newly introduced roots appear with modulo guards, e.g. Mod(dx, 2) = 0 suggests dx is a derived dim equal to 2 * _dx, introducing a new root _dx. Currently the final suggested fixes handle this correctly, but we can get intermediate results that look like {"dy": {"eq": "dx + 1"}, "dx": {"eq": "2 * _dx + 1, "min": 3, "max": 15}}, and this prettifies the suggested fixes routine by unifying to a single root, and making each intermediate suggestion either a derived dim or min/max range, not both. 2) The current suggested fixes for derived dims can lead to root dims/derived dims being swapped, e.g. `dy - 1, dy` -> `dx, dx + 1`. This leads to problematic suggested fixes that look like `dy - 1 = Dim("dy - 1")` since we don't have access to the original variable name. This PR only adds a suggested fix for the root dim, and removes all other derived suggestions. For example, with the export test case test_derived_dim_out_of_order_simplified: ``` _dimz = torch.export.Dim("_dimz", min=6, max=8) dimy = _dimz - 1 dimx = dimy - 1 dimz = torch.export.Dim("dimz", min=6, max=8) # doesn't work, should be = _dimz class Foo(torch.nn.Module): def forward(self, x, y, z): return x + y[1:] + z[2:] foo = Foo() u, v, w = torch.randn(5), torch.randn(6), torch.randn(7) export( foo, (u, v, w), dynamic_shapes=({0: dimx}, {0: dimy}, {0: dimz}), ) ``` Before: ``` Suggested fixes: _dimz = Dim('_dimz', min=3, max=9223372036854775807) # 2 <= _dimz - 1 <= 9223372036854775806 _dimz - 2 = Dim('_dimz - 2', min=4, max=6) _dimz = Dim('_dimz', min=2, max=9223372036854775806) # 2 <= _dimz <= 9223372036854775806 _dimz - 1 = _dimz - 1 dimz = _dimz ``` New suggested fixes: ``` Suggested fixes: dimz = _dimz ``` Note: This assumes the specified derived relations between dims are correct. This should be valid because: 1) if the relation is plain wrong (e.g. (dx, dx - 1) provided with inputs (6, 4)), this gets caught in beforehand in produce_guards. 2) if the relation is correct but does not match the emitted guard, for example: ``` def forward(self, x, y): return x.reshape([-1]) + y # guard: s0 * 2 = s1 dx = Dim("dx") export( model, (torch.randn(6, 2), torch.randn(12)), dynamic_shapes={"x": (dx, 2), "y": (dx + 6, )} ) ``` This produces two linear equations, leading to specialization since a) produce_guards is able to solve for a concrete value, and b) the export constraint solver will anyways force specializations due to range constraints. Test Plan: export test cases Differential Revision: D56974543

facebook-github-bot · 2024-05-09T05:58:07Z

This pull request was exported from Phabricator. Differential Revision: D56974543

pianpwk · 2024-05-09T06:12:11Z

torch/fx/experimental/symbolic_shapes.py

+                root = list(c["eq"].free_symbols)[0]
+                if not str(root) in name_to_dim:
+                    introduced_roots[str(root)] = k
+                    # calculate necessary min & max


moved this code from prettify_results, existed for newly introduced roots

pianpwk · 2024-05-09T06:13:31Z

torch/fx/experimental/symbolic_shapes.py

                    if isinstance(other, int):
                        others.append(f"{k} = {other}")
                    elif self._is_supported_equivalence(other):
-                        s = next(iter(other.free_symbols))


this handles newly introduced roots, move to new function

avikchaudhuri

Trusting the tests and the summary, the changes are great! Mostly skimmed over the code.

pianpwk · 2024-05-24T22:03:47Z

Trusting the tests and the summary, the changes are great! Mostly skimmed over the code.

Thank you!

facebook-github-bot · 2024-05-24T22:25:50Z

This pull request was exported from Phabricator. Differential Revision: D56974543

facebook-github-bot · 2024-05-25T00:54:30Z

This pull request was exported from Phabricator. Differential Revision: D56974543

…ixes (#125543) Summary: This PR address 2 issues with derived dim suggested fixes, 1) newly introduced roots, and 2) root swapping. 1 | Newly introduced roots appear with modulo guards, e.g. Mod(dx, 2) = 0 suggests dx is a derived dim equal to 2 * _dx, introducing a new root _dx. Currently the final suggested fixes handle this correctly, but we can get intermediate results where related derived dims don't rely on a unified root, and are a mixture of min/max range and derived suggestions. For example: ``` "dx": {"eq": 3*_dx-1, "max": 35} "dy": {"eq": dx+1} This should lead to suggested fixes _dx = Dim('_dx', max=12) dx = 3 * _dx - 1 dy = 3 * _dx ``` This PR prettifies the suggested fixes routine by unifying to a single root, and making each intermediate suggestion either a derived dim or min/max range, not both. 2 | The current suggested fixes for derived dims can lead to root dims/derived dims being swapped, e.g. `dy - 1, dy` -> `dx, dx + 1`. This leads to problematic suggested fixes that look like `dy - 1 = Dim("dy - 1")` since we don't have access to the original variable name. This PR only adds a suggested fix for the root dim, and removes all other derived suggestions. For example, with the export test case test_derived_dim_out_of_order_simplified: ``` _dimz = torch.export.Dim("_dimz", min=6, max=8) dimy = _dimz - 1 dimx = dimy - 1 dimz = torch.export.Dim("dimz", min=6, max=8) # doesn't work, should be = _dimz class Foo(torch.nn.Module): def forward(self, x, y, z): return x + y[1:] + z[2:] foo = Foo() u, v, w = torch.randn(5), torch.randn(6), torch.randn(7) export( foo, (u, v, w), dynamic_shapes=({0: dimx}, {0: dimy}, {0: dimz}), ) ``` Before: ``` Suggested fixes: _dimz = Dim('_dimz', min=3, max=9223372036854775807) # 2 <= _dimz - 1 <= 9223372036854775806 _dimz - 2 = Dim('_dimz - 2', min=4, max=6) _dimz = Dim('_dimz', min=2, max=9223372036854775806) # 2 <= _dimz <= 9223372036854775806 _dimz - 1 = _dimz - 1 dimz = _dimz ``` New suggested fixes: ``` Suggested fixes: dimz = _dimz ``` Note: This assumes the specified derived relations between dims are correct. This should be valid because: 1) if the relation is plain wrong (e.g. (dx, dx - 1) provided with inputs (6, 4)), this gets caught in beforehand in produce_guards. 2) if the relation is correct but does not match the emitted guard, for example: ``` def forward(self, x, y): return x.reshape([-1]) + y # guard: s0 * 2 = s1 dx = Dim("dx") export( model, (torch.randn(6, 2), torch.randn(12)), dynamic_shapes={"x": (dx, 2), "y": (dx + 6, )} ) ``` This produces two linear equations, leading to specialization since a) produce_guards is able to solve for a concrete value, and b) the export constraint solver will anyways force specializations due to range constraints. Test Plan: export test cases Reviewed By: avikchaudhuri Differential Revision: D56974543

facebook-github-bot · 2024-05-25T00:54:46Z

This pull request was exported from Phabricator. Differential Revision: D56974543

facebook-github-bot · 2024-05-28T16:15:15Z

This pull request was exported from Phabricator. Differential Revision: D56974543

facebook-github-bot · 2024-05-28T16:16:50Z

This pull request was exported from Phabricator. Differential Revision: D56974543

facebook-github-bot · 2024-05-28T16:53:12Z

This pull request was exported from Phabricator. Differential Revision: D56974543

…ixes (#125543) Summary: This PR address 2 issues with derived dim suggested fixes, 1) newly introduced roots, and 2) root swapping. 1 | Newly introduced roots appear with modulo guards, e.g. Mod(dx, 2) = 0 suggests dx is a derived dim equal to 2 * _dx, introducing a new root _dx. Currently the final suggested fixes handle this correctly, but we can get intermediate results where related derived dims don't rely on a unified root, and are a mixture of min/max range and derived suggestions. For example: ``` "dx": {"eq": 3*_dx-1, "max": 35} "dy": {"eq": dx+1} This should lead to suggested fixes _dx = Dim('_dx', max=12) dx = 3 * _dx - 1 dy = 3 * _dx ``` This PR prettifies the suggested fixes routine by unifying to a single root, and making each intermediate suggestion either a derived dim or min/max range, not both. 2 | The current suggested fixes for derived dims can lead to root dims/derived dims being swapped, e.g. `dy - 1, dy` -> `dx, dx + 1`. This leads to problematic suggested fixes that look like `dy - 1 = Dim("dy - 1")` since we don't have access to the original variable name. This PR only adds a suggested fix for the root dim, and removes all other derived suggestions. For example, with the export test case test_derived_dim_out_of_order_simplified: ``` _dimz = torch.export.Dim("_dimz", min=6, max=8) dimy = _dimz - 1 dimx = dimy - 1 dimz = torch.export.Dim("dimz", min=6, max=8) # doesn't work, should be = _dimz class Foo(torch.nn.Module): def forward(self, x, y, z): return x + y[1:] + z[2:] foo = Foo() u, v, w = torch.randn(5), torch.randn(6), torch.randn(7) export( foo, (u, v, w), dynamic_shapes=({0: dimx}, {0: dimy}, {0: dimz}), ) ``` Before: ``` Suggested fixes: _dimz = Dim('_dimz', min=3, max=9223372036854775807) # 2 <= _dimz - 1 <= 9223372036854775806 _dimz - 2 = Dim('_dimz - 2', min=4, max=6) _dimz = Dim('_dimz', min=2, max=9223372036854775806) # 2 <= _dimz <= 9223372036854775806 _dimz - 1 = _dimz - 1 dimz = _dimz ``` New suggested fixes: ``` Suggested fixes: dimz = _dimz ``` Note: This assumes the specified derived relations between dims are correct. This should be valid because: 1) if the relation is plain wrong (e.g. (dx, dx - 1) provided with inputs (6, 4)), this gets caught in beforehand in produce_guards. 2) if the relation is correct but does not match the emitted guard, for example: ``` def forward(self, x, y): return x.reshape([-1]) + y # guard: s0 * 2 = s1 dx = Dim("dx") export( model, (torch.randn(6, 2), torch.randn(12)), dynamic_shapes={"x": (dx, 2), "y": (dx + 6, )} ) ``` This produces two linear equations, leading to specialization since a) produce_guards is able to solve for a concrete value, and b) the export constraint solver will anyways force specializations due to range constraints. Test Plan: export test cases Reviewed By: avikchaudhuri Differential Revision: D56974543

facebook-github-bot · 2024-05-28T16:55:34Z

This pull request was exported from Phabricator. Differential Revision: D56974543

facebook-github-bot · 2024-05-28T20:39:53Z

@pytorchbot merge -f 'Landed internally'

(Initiating merge automatically since Phabricator Diff has merged, using force because this PR might not pass merge_rules.json but landed internally)

pytorchmergebot · 2024-05-28T20:41:32Z

Merge started

Your change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use -f as last resort and instead consider -i/--ignore-current to continue the merge ignoring current failures. This will allow currently pending tests to finish and report signal before the merge.

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

…ixes (pytorch#125543) Pull Request resolved: pytorch#125543 This PR address 2 issues with derived dim suggested fixes, 1) newly introduced roots, and 2) root swapping. 1 | Newly introduced roots appear with modulo guards, e.g. Mod(dx, 2) = 0 suggests dx is a derived dim equal to 2 * _dx, introducing a new root _dx. Currently the final suggested fixes handle this correctly, but we can get intermediate results where related derived dims don't rely on a unified root, and are a mixture of min/max range and derived suggestions. For example: ``` "dx": {"eq": 3*_dx-1, "max": 36} "dy": {"eq": dx+1} This should lead to suggested fixes _dx = Dim('_dx', max=12) dx = 3 * _dx - 1 dy = 3 * _dx ``` This PR prettifies the suggested fixes routine by unifying to a single root, and making each intermediate suggestion either a derived dim or min/max range, not both. 2 | The current suggested fixes for derived dims can lead to root dims/derived dims being swapped, e.g. `dy - 1, dy` -> `dx, dx + 1`. This leads to problematic suggested fixes that look like `dy - 1 = Dim("dy - 1")` since we don't have access to the original variable name. This PR only adds a suggested fix for the root dim, and removes all other derived suggestions. For example, with the export test case test_derived_dim_out_of_order_simplified: ``` _dimz = torch.export.Dim("_dimz", min=6, max=8) dimy = _dimz - 1 dimx = dimy - 1 dimz = torch.export.Dim("dimz", min=6, max=8) # doesn't work, should be = _dimz class Foo(torch.nn.Module): def forward(self, x, y, z): return x + y[1:] + z[2:] foo = Foo() u, v, w = torch.randn(5), torch.randn(6), torch.randn(7) export( foo, (u, v, w), dynamic_shapes=({0: dimx}, {0: dimy}, {0: dimz}), ) ``` Before: ``` Suggested fixes: _dimz = Dim('_dimz', min=3, max=9223372036854775807) # 2 <= _dimz - 1 <= 9223372036854775806 _dimz - 2 = Dim('_dimz - 2', min=4, max=6) _dimz = Dim('_dimz', min=2, max=9223372036854775806) # 2 <= _dimz <= 9223372036854775806 _dimz - 1 = _dimz - 1 dimz = _dimz ``` New suggested fixes: ``` Suggested fixes: dimz = _dimz ``` Note: This assumes the specified derived relations between dims are correct. This should be valid because: 1) if the relation is plain wrong (e.g. (dx, dx - 1) provided with inputs (6, 4)), this gets caught in beforehand in produce_guards. 2) if the relation is correct but does not match the emitted guard, for example: ``` def forward(self, x, y): return x.reshape([-1]) + y # guard: s0 * 2 = s1 dx = Dim("dx") export( model, (torch.randn(6, 2), torch.randn(12)), dynamic_shapes={"x": (dx, 2), "y": (dx + 6, )} ) ``` This produces two linear equations, leading to specialization since a) produce_guards is able to solve for a concrete value, and b) the export constraint solver will anyways force specializations due to range constraints. Pull Request resolved: pytorch#125543 Approved by: https://github.com/avikchaudhuri

pianpwk requested review from avikchaudhuri, gmagogsfm, tugsbayasgalan and zhxchen17 as code owners May 5, 2024 00:37

pytorch-bot bot added ciflow/inductor module: dynamo release notes: fx release notes category labels May 5, 2024

facebook-github-bot added the fb-exported label May 5, 2024

facebook-github-bot force-pushed the export-D56974543 branch from 39fefbf to 4d03382 Compare May 5, 2024 05:35

pianpwk force-pushed the export-D56974543 branch from 4d03382 to 35b2f34 Compare May 6, 2024 16:04

avikchaudhuri reviewed May 6, 2024

View reviewed changes

facebook-github-bot force-pushed the export-D56974543 branch from 35b2f34 to 6d82012 Compare May 9, 2024 05:57

pianpwk changed the title ~~[export] handle root swapping in derived dims suggested fixes~~ [export] handle new roots & root swapping in derived dims suggested fixes May 9, 2024

pianpwk commented May 9, 2024

View reviewed changes

avikchaudhuri approved these changes May 24, 2024

View reviewed changes

facebook-github-bot force-pushed the export-D56974543 branch from 6d82012 to 2c3f918 Compare May 24, 2024 22:25

facebook-github-bot force-pushed the export-D56974543 branch from 2c3f918 to 1825eda Compare May 25, 2024 00:54

pianpwk force-pushed the export-D56974543 branch from 1825eda to aaa0105 Compare May 25, 2024 00:54

facebook-github-bot force-pushed the export-D56974543 branch from aaa0105 to 1c29db7 Compare May 28, 2024 16:14

pianpwk force-pushed the export-D56974543 branch from 1c29db7 to 0b250f1 Compare May 28, 2024 16:16

facebook-github-bot force-pushed the export-D56974543 branch from 0b250f1 to b24262e Compare May 28, 2024 16:52

pianpwk force-pushed the export-D56974543 branch from b24262e to aff51b8 Compare May 28, 2024 16:55

pytorchmergebot added the merging label May 28, 2024

pytorchmergebot closed this in f206c5c May 28, 2024

pytorchmergebot added Merged and removed merging labels May 28, 2024

github-actions bot deleted the export-D56974543 branch June 28, 2024 01:56

[export] handle new roots & root swapping in derived dims suggested fixes #125543

[export] handle new roots & root swapping in derived dims suggested fixes #125543

Uh oh!

Conversation

pianpwk commented May 5, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot bot commented May 5, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/125543

✅ You can merge normally! (1 Unrelated Failure)

Uh oh!

facebook-github-bot commented May 5, 2024

Uh oh!

facebook-github-bot commented May 5, 2024

Uh oh!

facebook-github-bot commented May 6, 2024

Uh oh!

avikchaudhuri May 6, 2024

Choose a reason for hiding this comment

Uh oh!

avikchaudhuri May 6, 2024

Choose a reason for hiding this comment

Uh oh!

avikchaudhuri May 6, 2024

Choose a reason for hiding this comment

Uh oh!

avikchaudhuri May 6, 2024

Choose a reason for hiding this comment

Uh oh!

pianpwk May 7, 2024

Choose a reason for hiding this comment

Uh oh!

facebook-github-bot commented May 9, 2024

Uh oh!

pianpwk May 9, 2024

Choose a reason for hiding this comment

Uh oh!

pianpwk May 9, 2024

Choose a reason for hiding this comment

Uh oh!

avikchaudhuri left a comment

Choose a reason for hiding this comment

Uh oh!

pianpwk commented May 24, 2024

Uh oh!

facebook-github-bot commented May 24, 2024

Uh oh!

facebook-github-bot commented May 25, 2024

Uh oh!

facebook-github-bot commented May 25, 2024

Uh oh!

facebook-github-bot commented May 28, 2024

Uh oh!

facebook-github-bot commented May 28, 2024

Uh oh!

facebook-github-bot commented May 28, 2024

Uh oh!

facebook-github-bot commented May 28, 2024

Uh oh!

facebook-github-bot commented May 28, 2024

Uh oh!

pytorchmergebot commented May 28, 2024

Merge started

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

pianpwk commented May 5, 2024 •

edited

Loading

pytorch-bot bot commented May 5, 2024 •

edited

Loading