The model de#corporates a compact architecture utilizing depthwise separable convolutions to minimize parameters and FLOPs, inverted residual blocks (inspired by MobileNetV2) to balance depth and width efficiently, and channel reduction techniques to eliminate redundancy while maintaining expressiveness.
-
Notifications
You must be signed in to change notification settings - Fork 0
The model de#corporates a compact architecture utilizing depthwise separable convolutions to minimize parameters and FLOPs, inverted residual blocks (inspired by MobileNetV2) to balance depth and width efficiently, and channel reduction techniques.But the model has not reached target of 95% accuracy,I invite other hackers to try. hack
License
ianwolf99/CIFAR10-model-contest
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
The model de#corporates a compact architecture utilizing depthwise separable convolutions to minimize parameters and FLOPs, inverted residual blocks (inspired by MobileNetV2) to balance depth and width efficiently, and channel reduction techniques.But the model has not reached target of 95% accuracy,I invite other hackers to try. hack
Topics
Resources
License
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published