microsoft / superbenchmark

A validation and profiling tool for AI infrastructure
https://aka.ms/superbench
MIT License
270 stars 59 forks source link

Revise Code - Fix wrong torch usage in communication wrapper for Distributed Inference Benchmark #505

Closed yzygitzh closed 1 year ago

yzygitzh commented 1 year ago

Description This commit fixes wrong torch.empty_like usage and missing dtype and device argument in communication wrappers.

codecov[bot] commented 1 year ago

Codecov Report

Merging #505 (e611e2c) into release/0.8 (9d250cd) will not change coverage. The diff coverage is 0.00%.

@@             Coverage Diff              @@
##           release/0.8     #505   +/-   ##
============================================
  Coverage        87.31%   87.31%           
============================================
  Files               89       89           
  Lines             5944     5944           
============================================
  Hits              5190     5190           
  Misses             754      754           
Flag Coverage Δ
cpu-python3.6-unit-test 73.58% <0.00%> (ø)
cpu-python3.7-unit-test 73.58% <0.00%> (ø)
cpu-python3.8-unit-test 74.05% <0.00%> (ø)
cuda-unit-test 87.24% <0.00%> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files Coverage Δ
...ench/benchmarks/micro_benchmarks/dist_inference.py 81.03% <0.00%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.