-
Notifications
You must be signed in to change notification settings - Fork 240
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
Use convert_model in tests #2102
Use convert_model in tests #2102
Conversation
Codecov Report
Additional details and impacted files@@ Coverage Diff @@
## develop #2102 +/- ##
============================================
+ Coverage 0 36.16% +36.16%
============================================
Files 0 476 +476
Lines 0 42494 +42494
============================================
+ Hits 0 15369 +15369
- Misses 0 27125 +27125
|
tests/post_training/requirements.txt
Outdated
@@ -7,7 +7,7 @@ onnx==1.13.1 | |||
onnxruntime==1.14.1 | |||
pytest | |||
pytest-cov | |||
openvino-dev==2023.0.1 | |||
openvino-dev==2023.1.0.dev20230811 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why was it changed? Do we have any proof that this package is not buggy?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
2023.1.0.dev20230811 have some bug that was fixed on current release branch.
But with 2023.0.1 mo.convert_model does not support quantized models.
Until dump new openvino package it's the best option.
6271a41
to
bf25a3b
Compare
post_training_quantization/149/ |
Update e2e test will be done in another PR. |
post_training_quantization/152 - with dump fp32 models |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What about PyTorch and TensorFlow backends?
@@ -9,7 +9,7 @@ | |||
# See the License for the specific language governing permissions and | |||
# limitations under the License. | |||
|
|||
__version__ = "2.5.0" | |||
__version__ = "2.6.0" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@KodiaqQ, I suggest to merge this PR after NNCF 2.6.0 release.
Changes
Use mo.convert_model instead of converting via onnx in tests.
nncf.__version__ >= 2.6.0
inception_resnet_v2, levit_128, vgg11
in test_quantize_conformance.py (120012)Reason for changes
New api in openvino to convert PT models
Related tickets
117885