NGC | Catalog
Welcome Guest
CatalogModelsResNeXt101-32x4d pretrained weights (PyTorch, AMP, ImageNet)

ResNeXt101-32x4d pretrained weights (PyTorch, AMP, ImageNet)

For downloads and more information, please view on a desktop device.
Logo for ResNeXt101-32x4d pretrained weights (PyTorch, AMP, ImageNet)

Description

ResNeXt101-32x4d ImageNet pretrained weights

Publisher

NVIDIA Deep Learning Examples

Use Case

Classification

Framework

PyTorch

Latest Version

20.06.0

Modified

October 29, 2021

Size

169.15 MB

Model Overview

ResNet with bottleneck 3x3 Convolutions substituted by 3x3 Grouped Convolutions.

Model Architecture

ResNextArch

Image source: Aggregated Residual Transformations for Deep Neural Networks

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.

Training

This model was trained using script available on NGC and in GitHub repo

Dataset

The following datasets were used to train this model:

  • ImageNet - Image database organized according to the WordNet hierarchy, in which each noun is depicted by hundreds and thousands of images.

Performance

Performance numbers for this model are available in NGC

References

License

This model was trained using open-source software available in Deep Learning Examples repository. For terms of use, please refer to the license of the script and the datasets the model was derived from.