huggingface / huggingface/optimum-amd

[RyzenAI] Operator regression on Ryzen SDK 1.1

Open
#124 0 comments 0 reactions 0 assignees View on GitHub
Ryzen AI
Dominant language
Jupyter Notebook
Stars
101
Forks
21
Avg merge
58m
Merged PRs (30d)
1

Description

The following models have shown regression:

The format is: `Ryzen1.1 value (Ryzen1.0.1 value)`

```
Total Ops | DPU Ops | CPU Ops | Reg. | Model
944 | 296(331) | 648(613) | Y | timm/eca_botnext26ts_256.c1_in1k
1212 | 1042(1126) | 170(86) | Y | timm/eca_nfnet_l0.ra2_in1k
755 | 613(683) | 142(72) | Y | timm/eca_resnet33ts.ra2_in1k
611 | 497(553) | 114(58) | Y | timm/eca_resnext26ts.ch_in1k
431 | 317(373) | 114(58) | Y | timm/ecaresnet26t.ra2_in1k
1325 | 1078(1131) | 247(194) | Y | timm/gcresnet33ts.ra2_in1k
1059 | 889(921) | 170(138) | Y | timm/gcresnext26ts.ch_in1k
```

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.