Summary:
This pr fixes all the places in strict export stack where the output node's meta is not preserved correctly. However, we're getting a new error for the test we intend to fix: `buck2 run caffe2/test/quantization:test_quantization -- -r "test_re_export_preserve_handle"`:
The `get_attr` nodes has wrong metadata. I guess there are more things need to be fixed to get it working but it's beyond the scope of this PR.
Test Plan: buck2 run caffe2/test/quantization:test_quantization -- -r "test_re_export_preserve_handle"
Differential Revision: D60198221
Pull Request resolved: https://github.com/pytorch/pytorch/pull/131706
Approved by: https://github.com/yushangdi
Summary: Fix index issues in torch.fx.interpreter by changing range from `[:i]` to `[:i+1]`. Because if there are `n` elements, the last index `i` of the `for` loop is `n-1` and `[:i]` can only get access to elements from index `0` to index `n-2` and miss the last element. `[:i+1]` can get access to all elements correctly.
Test Plan: Test with Node API
Differential Revision: D59028395
Pull Request resolved: https://github.com/pytorch/pytorch/pull/129527
Approved by: https://github.com/dulinriley
I feel it's easier to open a new PR rather than iterating on the previous PR (https://github.com/pytorch/pytorch/pull/105257 ) since this is more like a rewrite.
In this PR, instead of changing GraphModule directly which can easily causes BC issue, I create a LazyGraphModule class as Zachary & Jason suggested in comments from the previous PR.
The difference between LazyGraphModule and GraphModule is mainly about how re-compile for the graph module happens. In GraphModule the recompilation happens 'eagerly': constructing a GraphModule will cause the recompilation. While in LazyGraphModule, we just mark the module as needing recompilation. The real recompilation only happens when absolutely required (e.g. call forward method, access the code property etc.). In a lot of cases in torch.compile, the real recompilation eventually is not triggered at all. This can save a few seconds of compilation time.
By default, GraphModule rather than LazyGraphModule is used. `use_lazy_graph_module(True)` context manager can be used to pick LazyGraphModule instead. This has been applied to the torch.compile stack.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/117911
Approved by: https://github.com/jansel
UnflattenedModule doesn't really behave like a graph module; we customize `__call__` to do something completely different than what GraphModule does. So, things that test `isinstance(unflattened_module, GraphModule)` and do something with the GraphModule are often broken.
This change makes UnflattenedModule it's own thing.
Differential Revision: [D51959097](https://our.internmc.facebook.com/intern/diff/D51959097/)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/115408
Approved by: https://github.com/zhxchen17
This is a two part PR; I can split it if you really want me to.
The first part is a refactor of the after aot repro/minifier scripts to come with a command line interface. I maintain exact BC with the previous interface (so, e.g., you still get a repro.py and a run_minifier.py that do the same thing as before), but each of these scripts also take command line arguments now which you can use to customize what actually happens. Check `run_repro` for full documentation on the arguments.
The second part of this is an implementation of `analyze` subcommand on the new CLI for any repro.
<img width="1277" alt="image" src="https://user-images.githubusercontent.com/13564/235045677-8545aab7-5e83-4813-bbec-47783dc60122.png">
This facility is oriented towards accuracy debugging. It does several things:
1. It will run your model twice and check for nondeterminism in inductor/float64, *even* on intermediate inputs (our benchmarking nondeterminism test only checks for nondeterminism on the final output). This makes localizing which operator is nondeterministic easy.
2. It will run your compiled model side-by-side with eager and float64 variants, and then report when things diverge too far from RMSE delta from float64.
Importantly, it does all this without requiring every intermediate to be held in memory (which will cause an OOM on large repros, such as the one I tested this on.)
Some other minor improvements:
* MinifierTestBase now has an easy to comment out spot that you can use to retain the temporary directory; good for debugging
* We print "running minifier" and "running repro" in MinifierTestBase to make it easier to orient where logs are coming from
* same takes a `log_error` optional argument which you can use to reroute the error logs when things mismatch
* counters["inductor"]["intermediate_hooks"] tracks the number of intermediate hooks we've codegen'ed; good for populate the tqdm interface
* torch.fx.interpreter gets an official `boxed_run` interface which uses the boxed arguments calling convention and doesn't retain inputs unnecessarily long
* torch.utils._content_store gets compute_tensor_metadata/read_tensor_metadata helper functions for computing tensor information without serializing it
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100226
Approved by: https://github.com/bertmaher, https://github.com/bdhirsh, https://github.com/anijain2305
Builds up sympy expressions computing the lower and upper bound of ranges, and then finds `op.to_dtype(x, torch.int64)` nodes whose dominated values can all be computed in a lower precision. I haven't gotten all the way to work with dynamic shapes but it should be a fairly small change. There's still additional work to get torchinductor to work with large tensors (see https://github.com/pytorch/torchdynamo/issues/1819) because we would need to add explicit dtype annotations to int64 which we're not doing right now.
Fix for https://github.com/pytorch/torchdynamo/issues/1293.
Performance Before OpBench aten.upsample_bilinear2d.vec float32:
(25th %, 50th %, 75th %)
Before
[0.7521964636710751, 0.8645357996607477, 2.8746003906598494]
After:
[0.9511363478204263, 1.0295566597806718, 3.2662165264101755]
Pull Request resolved: https://github.com/pytorch/pytorch/pull/91028
Approved by: https://github.com/jansel
This commit had inconsistent internal land and pr merged. This caused merge conflicts that required revert in both places, normalize the internal commit stack, and then re-land properly.
Original commit: #88384 (011452a2a1)
Inconsistent revert: #90018 (8566aa7c0b4bdca50bf85ca14705b4304de030b3)
Revert of the inconsistent revert to restore healthy state (or re-land of the original commit): cf3c3f2280
Landing the correct, internally congruent revert of the original commit: (This PR) #90055 (TBD)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/90055
Approved by: https://github.com/DanilBaibak, https://github.com/malfet
We have been using this extra error context in the symbolic-shapes
branch and it is quite useful. Contributing it upstream here.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/85810
Approved by: https://github.com/albanD
