Skip to content
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.

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

Improvements to propagate_const for int4 #3494

Open
wants to merge 9 commits into
base: develop
Choose a base branch
from
Open

Conversation

pfultz2
Copy link
Collaborator

@pfultz2 pfultz2 commented Oct 1, 2024

No description provided.

@pfultz2 pfultz2 requested a review from causten as a code owner October 1, 2024 17:10
@migraphx-bot
Copy link
Collaborator

Test Batch Rate new
23bd0d
Rate old
9fb2f4
Diff Compare
torchvision-resnet50 64 3,255.85 3,258.31 -0.08%
torchvision-resnet50_fp16 64 6,969.83 6,980.08 -0.15%
torchvision-densenet121 32 2,431.87 2,430.13 0.07%
torchvision-densenet121_fp16 32 4,084.00 4,100.79 -0.41%
torchvision-inceptionv3 32 1,635.64 1,635.80 -0.01%
torchvision-inceptionv3_fp16 32 2,754.30 2,756.32 -0.07%
cadene-inceptionv4 16 778.80 780.08 -0.16%
cadene-resnext64x4 16 807.04 807.36 -0.04%
slim-mobilenet 64 7,464.04 7,463.42 0.01%
slim-nasnetalarge 64 208.08 208.23 -0.07%
slim-resnet50v2 64 3,438.00 3,435.41 0.08%
bert-mrpc-onnx 8 1,146.54 1,146.86 -0.03%
bert-mrpc-tf 1 314.31 314.12 0.06%
pytorch-examples-wlang-gru 1 394.33 422.52 -6.67% 🔴
pytorch-examples-wlang-lstm 1 380.20 390.72 -2.69%
torchvision-resnet50_1 1 808.71 811.85 -0.39%
cadene-dpn92_1 1 398.15 435.19 -8.51% 🔴
cadene-resnext101_1 1 380.70 380.85 -0.04%
onnx-taau-downsample 1 366.31 366.80 -0.14%
dlrm-criteoterabyte 1 35.07 35.02 0.12%
dlrm-criteoterabyte_fp16 1 58.13 58.22 -0.15%
agentmodel 1 8,286.17 7,495.47 10.55% 🔆
unet_fp16 2 58.99 58.91 0.13%
resnet50v1_fp16 1 935.63 943.39 -0.82%
resnet50v1_int8 1 966.59 969.89 -0.34%
bert_base_cased_fp16 64 1,170.82 1,171.87 -0.09%
bert_large_uncased_fp16 32 363.13 363.41 -0.08%
bert_large_fp16 1 211.08 213.38 -1.08%
distilgpt2_fp16 16 2,201.22 2,202.98 -0.08%
yolov5s 1 545.14 541.76 0.62%
tinyllama 1 43.44 43.64 -0.46%
vicuna-fastchat 1 171.70 172.03 -0.19%
whisper-tiny-encoder 1 420.21 417.60 0.62%
whisper-tiny-decoder 1 436.36 423.58 3.02% 🔆

This build is not recommended to merge 🔴

@migraphx-bot
Copy link
Collaborator


     ✅ bert-mrpc-onnx: PASSED: MIGraphX meets tolerance

     ✅ bert-mrpc-tf: PASSED: MIGraphX meets tolerance

     ✅ pytorch-examples-wlang-gru: PASSED: MIGraphX meets tolerance

     ✅ pytorch-examples-wlang-lstm: PASSED: MIGraphX meets tolerance

     ✅ torchvision-resnet50_1: PASSED: MIGraphX meets tolerance

     ✅ cadene-dpn92_1: PASSED: MIGraphX meets tolerance

     ✅ cadene-resnext101_1: PASSED: MIGraphX meets tolerance

     ✅ dlrm-criteoterabyte: PASSED: MIGraphX meets tolerance

     ✅ agentmodel: PASSED: MIGraphX meets tolerance

     ✅ unet: PASSED: MIGraphX meets tolerance

     ✅ resnet50v1: PASSED: MIGraphX meets tolerance

     ✅ bert_base_cased_fp16: PASSED: MIGraphX meets tolerance

🔴bert_large_uncased_fp16: FAILED: MIGraphX is not within tolerance - check verbose output


     ✅ bert_large: PASSED: MIGraphX meets tolerance

     ✅ yolov5s: PASSED: MIGraphX meets tolerance

     ✅ tinyllama: PASSED: MIGraphX meets tolerance

     ✅ vicuna-fastchat: PASSED: MIGraphX meets tolerance

     ✅ whisper-tiny-encoder: PASSED: MIGraphX meets tolerance

     ✅ whisper-tiny-decoder: PASSED: MIGraphX meets tolerance

     ✅ distilgpt2_fp16: PASSED: MIGraphX meets tolerance

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants