Skip to content

README.md

Latest commit

 

History

History
69 lines (48 loc) · 2.13 KB

README.md

File metadata and controls

69 lines (48 loc) · 2.13 KB
Raw

ConvNeXt Small (ixRT)

Model Description

The ConvNeXt Small model represents a significant stride in the evolution of convolutional neural networks (CNNs), introduced by researchers at Facebook AI Research (FAIR) and UC Berkeley. It is part of the ConvNeXt family, which challenges the dominance of Vision Transformers (ViTs) in the realm of visual recognition tasks.

Supported Environments

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

Model Preparation

Prepare Resources

Pretrained model: https://download.pytorch.org/models/convnext_small-0c510722.pth

Dataset: https://www.image-net.org/download.php to download the validation dataset.

Install Dependencies

# Install libGL
## CentOS
yum install -y mesa-libGL
## Ubuntu
apt install -y libgl1-mesa-glx

pip3 install -r ../../ixrt_common/requirements.txt

Model Conversion

mkdir checkpoints
python3 ../../ixrt_common/export.py --model-name convnext_small --weight convnext_small-0c510722.pth --output checkpoints/convnext_small.onnx

Model Inference

export PROJ_DIR=./
export DATASETS_DIR=/path/to/imagenet_val/
export CHECKPOINTS_DIR=./checkpoints
export RUN_DIR=../../ixrt_common/
export CONFIG_DIR=../../ixrt_common/config/CONVNEXT_SMALL_CONFIG

FP16

# Accuracy
bash scripts/infer_convnext_small_fp16_accuracy.sh
# Performance
bash scripts/infer_convnext_small_fp16_performance.sh

Model Results

Model BatchSize Precision FPS Top-1(%) Top-5(%)
ConvNeXt Small 32 FP16 323.508 83.302 96.548