onnxruntime/orttraining
Oliver Rausch 1685ab8138
Implement Concat with Strided copy (#8336)
Adds a StridedCopy function that implements a copy from strided tensor to another.

This parallelizes the Concat operator, and can also be used in the future to parallelize many other data movement operators (e.g. Transpose, Split, etc.).
This operation is also required for the proposed data layout extensions to ORT.
2021-07-27 18:27:56 +02:00
..
orttraining Implement Concat with Strided copy (#8336) 2021-07-27 18:27:56 +02:00
pytorch_frontend_examples Sync ORTModule branch with master and fix tests (#6526) 2021-02-02 08:59:56 -08:00
tools [ROCm] dockerfile updates (#7955) 2021-06-10 23:50:19 -07:00