This repository provides a non-official PyTorch reimplementation of the CBAM paper from ECCV 2018. CBAM is a plug-and-play attention module that can be integrated into any CNN architecture to adaptively refine intermediate feature maps. The implementation includes channel attention and spatial attention sub-modules applied sequentially at each convolutional block. Results are validated on ImageNet-1K using ResNet50, showing improved top-1 accuracy over baseline.