EfficientNet B4 is a member of the EfficientNet family, a series of convolutional neural network architectures that are designed to achieve excellent accuracy and efficiency. Introduced by researchers at Google, EfficientNets utilize the compound scaling method, which uniformly scales the depth, width, and resolution of the network to improve accuracy and efficiency.
| GPU | IXUCA SDK | Release | Branch |
|---|---|---|---|
| MR-V100 | 4.4.0 | 26.03 | release/26.03 |
| MR-V100 | 4.3.0 | 25.12 | release/25.12 |
Note: 请切换到与您的 SDK 版本对应的 Release 分支进行测试。请勿直接在 master 分支上运行测试，因为 master 分支可能包含与您的本地 SDK 版本不兼容的最新更改。
切换分支命令示例：
git checkout release/26.03
Pretrained model: https://download.pytorch.org/models/efficientnet_b4_rwightman-23ab8bcd.pth
Dataset: https://www.image-net.org/download.php to download the validation dataset.
pip3 install -r ../../ixrt_common/requirements.txtmkdir checkpoints
python3 ../../ixrt_common/export.py --model-name efficientnet_b4 --weight efficientnet_b4_rwightman-23ab8bcd.pth --output checkpoints/efficientnet_b4.onnxexport PROJ_DIR=./
export DATASETS_DIR=/path/to/imagenet_val/
export CHECKPOINTS_DIR=./checkpoints
export RUN_DIR=../../ixrt_common/
export CONFIG_DIR=../../ixrt_common/config/EFFICIENTNET_B4_CONFIG# Accuracy
bash scripts/infer_efficientnet_b4_fp16_accuracy.sh
# Performance
bash scripts/infer_efficientnet_b4_fp16_performance.sh| Model | BatchSize | Precision | FPS | Top-1(%) | Top-5(%) |
|---|---|---|---|---|---|
| EfficientNet_B4 | 32 | FP16 | 1204.94 | 79.27 | 94.53 |