File: cudnn_persistent_rnn.rst

package info (click to toggle)
pytorch 1.13.1%2Bdfsg-4
  • links: PTS, VCS
  • area: main
  • in suites: bookworm
  • size: 139,252 kB
  • sloc: cpp: 1,100,274; python: 706,454; ansic: 83,052; asm: 7,618; java: 3,273; sh: 2,841; javascript: 612; makefile: 323; xml: 269; ruby: 185; yacc: 144; objc: 68; lex: 44
file content (9 lines) | stat: -rw-r--r-- 305 bytes parent folder | download | duplicates (3) 
1
2
3
4
5
6
7
8
9
 
.. note::

    If the following conditions are satisfied:
    1) cudnn is enabled,
    2) input data is on the GPU
    3) input data has dtype ``torch.float16``
    4) V100 GPU is used,
    5) input data is not in ``PackedSequence`` format
    persistent algorithm can be selected to improve performance.