| --- |
| library_name: pytorch |
| --- |
| |
|  |
|
|
| EfficientNet-Lite is a family of mobile and IoT-friendly image classification models derived from EfficientNet and optimized for efficient inference on mobile CPUs, GPUs, and Edge TPUs, with modifications that improve quantization and hardware compatibility. |
|
|
| Original paper: [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://arxiv.org/abs/1905.11946) |
|
|
| # EfficientNet-Lite0 |
|
|
| This model uses the EfficientNet-Lite0 architecture, which removes squeeze-and-excitation modules, replaces Swish with ReLU6, and keeps the stem and head fixed across scaled variants to reduce latency and improve post-training quantization. |
|
|
| Model Configuration: |
| - Reference implementation: [timm.models.efficientnet_lite0](https://github.com/pprp/timm/blob/master/timm/models/efficientnet.py) |
| - Original Weight: [efficientnet_lite0.ra_in1k](https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_lite0_ra-37913777.pth) |
| - Dataset: [ImageNet](https://image-net.org) |
| - Resolution: 3x224x224 |
| - Support Cooper version: |
| - Cooper SDK: [2.5.4] |
| - Cooper Foundry: [2.3] |
|
|
| | Model | Device | compression | Model Link | |
| | :-----: | :-----: | :-----: | ------- | |
| | EfficientNet-Lite0 | N1-655 | Activation_fp16 | [Model_Link](https://huggingface.co/Ambarella/EfficientNet/blob/main/n1-655_efficientnet_lite0_act16.bin) | |
| | EfficientNet-Lite0 | N1-655 | Amba_optimized | [Model_Link](https://huggingface.co/Ambarella/EfficientNet/blob/main/n1-655_efficientnet_lite0_amba_optimized.bin) | |
| | EfficientNet-Lite0 | X7 | Activation_fp16 | [Model_Link](https://huggingface.co/Ambarella/EfficientNet/blob/main/x7_efficientnet_lite0_act16.bin) | |
| | EfficientNet-Lite0 | X7 | Amba_optimized | [Model_Link](https://huggingface.co/Ambarella/EfficientNet/blob/main/x7_efficientnet_lite0_amba_optimized.bin) | |
| | EfficientNet-Lite0 | CV7 | Activation_fp16 | [Model_Link](https://huggingface.co/Ambarella/EfficientNet/blob/main/cv7_efficientnet_lite0_act16.bin) | |
| | EfficientNet-Lite0 | CV7 | Amba_optimized | [Model_Link](https://huggingface.co/Ambarella/EfficientNet/blob/main/cv7_efficientnet_lite0_amba_optimized.bin) | |
| | EfficientNet-Lite0 | CV72 | Activation_fp16 | [Model_Link](https://huggingface.co/Ambarella/EfficientNet/blob/main/cv72_efficientnet_lite0_act16.bin) | |
| | EfficientNet-Lite0 | CV72 | Amba_optimized | [Model_Link](https://huggingface.co/Ambarella/EfficientNet/blob/main/cv72_efficientnet_lite0_amba_optimized.bin) | |
| | EfficientNet-Lite0 | CV75 | Activation_fp16 | [Model_Link](https://huggingface.co/Ambarella/EfficientNet/blob/main/cv75_efficientnet_lite0_act16.bin) | |
| | EfficientNet-Lite0 | CV75 | Amba_optimized | [Model_Link](https://huggingface.co/Ambarella/EfficientNet/blob/main/cv75_efficientnet_lite0_amba_optimized.bin) | |
|
|
