Skip to content

Update resnet/faster-rcnn (fp32) tuning logs #10

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
dsbarinov1 wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Update resnet/faster-rcnn (fp32) tuning logs #10

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -39,8 +39,8 @@ In the table below you can see the performance numbers for a set of models from
| TVM FP16 (VM) | 500,03| 63,63| 208,31| 3,11| 24,63|
| TVM FP16a32 (GE) | 403,74| 80,46| 260,57| 4,07| 29,6857|
| TVM FP16a32 (VM) | 662,15| 94,41| 265,45| 4,11| -|
| TVM FP32 (GE) | 496,67| 80,44| 317,15| 4,22| 35,46|
| TVM FP32 (VM) | 754,33| 95,14| 316,05| 4,28| 42,43|
| TVM FP32 (GE) | 350,85| 80,44| 273,22| 4,22| 35,46|
| TVM FP32 (VM) | 610,92| 95,14| 277,96| 4,28| 42,43|

*Virtual Machine has performance degradation only on Non Max Suppression layers, which seems to be a bug, that nevertheless affects the final performance ([issue#15405](https://github.com/apache/tvm/issues/15405)).*

Expand Down Expand Up @@ -172,4 +172,4 @@ profiling separate OpenCL kernels:
kernels on Android device and collect performance metrics.

- `OpenCLCPPRunner` is an native application which can be used to run OpenCL
kernels on Android device or on a host and collect performance metrics.
kernels on Android device or on a host and collect performance metrics.
Loading