ResNet with bottleneck 3x3 Convolutions substituted by 3x3 Grouped Convolutions.
Image shows difference between ResNet bottleneck block and ResNeXt bottleneck block.
ResNeXt101-32x4d model's cardinality equals to 32 and bottleneck width equals to 4.
The following datasets were used to train this model:
Performance numbers for this model are available in NGC