Aaron Orenstein

Results 26 issues of Aaron Orenstein

Tacotron2 causes massive loop unrolling resulting in very large graphs (26k nodes) which was causing inductor (and tracing itself) to choke. The unrolling size is controlled by the environment variable...

module: dynamo
ciflow/inductor

A small medley of fixes: - When validating sparse tensor indices don't check numel() if it's symbolic. - When validating sparse tensor indices if the indices are a FakeTensor then...

release notes: sparse
module: inductor
module: dynamo
ciflow/inductor

`Tensor.__repr__` calls functions which can perform logging which ends up logging `self` (with `__repr__`) causing an infinite loop. Detect this in `__repr__` and early-out instead of recursing. Another possible fix...

Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #124545

Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * #125312 * __->__ #124226 * #124225 * #124224 * #124223 * #122911

ciflow/inductor

Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * #125312 * #124226 * __->__ #124225 * #124224 * #124223 * #122911

ciflow/inductor

Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * #125312 * #124226 * #124225 * __->__ #124224 * #124223 * #122911

ciflow/inductor

When dispatching a fake tensor op we cache the result with `(op, args)` as the key. There are some args (such as one with a dynamic output shape) where the...

ciflow/inductor

When constructing a `FakeTensorMode`, instead of immediately formatting a full stack trace, grab the traceback and only format it on demand. 4.2% FakeTensor perf win on the microbenchmark. ``` import...

topic: not user facing
ciflow/inductor

We save and restore the DynamicLayerStack during frame eval but since fx graph has no way to express a try/finally we just assume it will happen. If we throw an...

module: dynamo
ciflow/inductor