ImageNet-1K(ILSVRC2012)|图像分类数据集|机器学习数据集
收藏NoisyNN 数据集概述
数据集
- ImageNet-1K (ILSVRC2012)
-
下载地址: ImageNet-1K
-
数据结构:
ImageNet1K/ ├── train/ │ ├── n01440764/ │ │ ├── n01440764_18.JPEG │ │ ├── n01440764_36.JPEG │ │ └── ... │ ├── n01443537/ │ └── ... │ └── n01484850/ ├── val/ │ ├── n01440764/ │ │ ├── ILSVRC2012_val_00000293.JPEG │ │ ├── ILSVRC2012_val_00002138.JPEG │ │ └── ... │ ├── n01443537/ │ └── ... │ └── n01484850/
-
数据预处理:
- 使用
unzip_tra.sh
和preprocess.py
进行数据预处理。 bash sh unzip_tra.sh python preprocess.py
- 使用
-
训练
-
训练命令示例:
python Main.py --lr 0.000001 --epochs 50 --batch_size 16 --layer 11 --gpu_id 0 --res 384 --patch_size 16 --scale base --noise_type linear --datasets ImageNet --num_classes 1000 --tra 0 --inf 1 --OptimalQ 1
-
训练命令详细说明可在
runScript.txt
中找到。
引用
-
NoisyNN:
@article{Yu2023NoisyNN, title={NoisyNN: Exploring the Impact of Information Entropy Change in Learning Systems}, author={Yu, Xiaowei and Huang, Zhe and Xue, Yao and Zhang, Lu and Wang, Li and Liu, Tianming and Dajiang Zhu}, journal={arXiv preprint arXiv:2309.10625}, year={2023} }
-
InterLUDE:
@article{Huang2024InterLUDE, title={InterLUDE: Interactions between Labeled and Unlabeled Data to Enhance Semi-Supervised Learning}, author={Huang, Zhe and Yu, Xiaowei and Zhu, Dajiang and Michael C. Hughes}, journal={International Conference on Machine Learning}, year={2024} }
-
FFTAT:
@article{Yu2025FFTAT, title={Feature Fusion Transferability Aware Transformer for Unsupervised Domain Adaptation}, author={Yu, Xiaowei and Huang, Zhe and Zao Zhang}, journal={IEEE/CVF Winter Conference on Applications of Computer Vision}, year={2025} }
