The New Nano Version of ConvNext is Awesome!
Latest release of timm (pytorch image models) has a new convnext variant that it's even better for fine-tuning.
Created on July 28|Last edited on July 30
Comment
Latest timm 0.6.7 release has a new variant of the ConvNext architecture that it's even faster and performs extremely well on fine-tuning and downstream tasks. This nano model is my new favorite for rapid iteration!
ConvNextNano vs other Small ConvNexts
The new nano variant against her siblings on the PETS datasets fine-tunning. You can read more about the methodology on the Finding the new Resnet18 in 2022 report.
Run set
24
Finding the New ResNet18
In this article, we try to understand how to choose the right backbone for fine-tuning a model on downstream datasets in an attempt to find the new ResNet18.
Leveraging Pre-Trained Models for Image Classification
In this article, we fine-tune a pre-trained model on a new classification dataset, to understand how well transfer learning helps the model train on new data.
Add a comment
Tags: ML News
Iterate on AI agents and models faster. Try Weights & Biases today.