-
Notifications
You must be signed in to change notification settings - Fork 212
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
NNCF v2.5.0 accuracy and performance degradation of EfficientDet #1935
Comments
@lisosia have you measured accuracy of initial (FP32/FP16) OpenVINO IR? Have you tried to obtain OpenVINO IR with the latest instruction? |
I attaches the zip file
No. |
@lisosia could you please try to use latest NNCF version? python ~/nncf/tests/openvino/tools/calibrate.py -c quantization_config.json --impl native
accuracy_check -c accuracy_check.yaml -m efficientdet-d0_quantization/efficientdet-d0_frozen.xml coco_precision: 31.33% python ~/nncf/tests/openvino/tools/calibrate.py -c quantization_config.json --impl pot
accuracy_check -c accuracy_check.yaml -m efficientdet-d0_quantization/efficientdet-d0_frozen.xml coco_precision: 30.87%
{
"compression": {
"algorithms": [
{
"name": "DefaultQuantization",
"params": {
"preset": "performance",
"stat_subset_size": 300
}
}
],
"dump_intermediate_model": true
},
"engine": {
"config": "<path to accuracy_check.yaml>"
},
"model": {
"model": "efficientdet-d0_frozen.xml",
"model_name": "efficientdet-d0",
"weights": "efficientdet-d0_frozen.bin"
}
} I couldn't reproduce accuracy drop with provided script too. |
@l-bat
I found
It's weird that FP32 accuracy is different from my result |
Note that I hope nncf (native-backend) will yield the same accuracy/throughput with pot-backend |
INT8 native: INT8 use_pot: |
Oh, I misread that. 31.33% was the int8(native) accuracy. BTW, I'll try nncf/tests/openvino/tools/calibrate.py tomorrow. |
FP32 accuracy is 31.93%. Why do you use |
@lisosia Could you please try to use advanced_parameters=AdvancedQuantizationParameters(overflow_fix="enable") ? |
Regarding dataset settings, I copied the following sample
I use Intel(R) Core(TM) i7-8700 CPU, which does not support VNNI I think.
overflow-fix="enable" didn't make significant effect. |
But you provided config with dataset: |
@l-bat |
@l-bat
We hope these problems will be corrected. |
@l-bat
|
I tried NNCF 2.5.0 for my effdet model and observed accuracy and performance degradation.
Even if I use pot-backend (
use_pot=True
), the accuracy and performance is degraded compared to nncf 2.4.0.The issue is reproduced with the efficientdet-d0 which was exported by the official procedure.
The throughput was measured by benchmark-app of openvino-dev=2023.0.0.
I attaches the reproduction code and environments.
I hope the issue will be fixed so that we can migrate to the newer NNCF.
repro_effdet_issue.zip
The text was updated successfully, but these errors were encountered: