mrq/bitsandbytes-rocm
1
0
Fork 1
Code Issues 1 Pull Requests Packages Projects Releases Wiki Activity
e67bfccbcd
bitsandbytes-rocm/speed_benchmark
History
Tim Dettmers ec1ea63711 Refactored triton into its own folder. Refactored fp8 matmuls.
2023-04-12 09:39:39 -07:00
..
info_a100_py2.jsonl cleaning and refactor 2023-04-01 18:46:04 +00:00
make_plot_with_jsonl.py clarify in readme 2023-04-01 23:50:12 +00:00
plot_with_info.pdf cleaning and refactor 2023-04-01 18:46:04 +00:00
README.md clarify in readme 2023-04-01 23:50:12 +00:00
speed_benchmark.py Refactored triton into its own folder. Refactored fp8 matmuls. 2023-04-12 09:39:39 -07:00

README.md

Steps:

  1. Run python speed_benchmark/speed_benchmark.py which times operations and writes their time to speed_benchmark/info_a100_py2.jsonl (change the name of the jsonl to a different name for your profiling).
  2. Run python speed_benchmark/make_plot_with_jsonl.py, which produces the speed_benchmark/plot_with_info.pdf. Again make sure you change the jsonl which is being processed.