facebookresearch / dropoutLinks
Code release for "Dropout Reduces Underfitting"
☆313Updated 2 years ago
Alternatives and similar repositories for dropout
Users that are interested in dropout are comparing it to the libraries listed below
Sorting:
- FFCV-SSL Fast Forward Computer Vision for Self-Supervised Learning.☆208Updated 2 years ago
- Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time☆477Updated last year
- Implementation of fused cosine similarity attention in the same style as Flash Attention☆214Updated 2 years ago
- Implementation of the Adan (ADAptive Nesterov momentum algorithm) Optimizer in Pytorch☆252Updated 2 years ago
- ☆185Updated last year
- Probing the representations of Vision Transformers.☆328Updated 2 years ago
- Implementation of Soft MoE, proposed by Brain's Vision team, in Pytorch☆309Updated 4 months ago
- Implementation of Discrete Key / Value Bottleneck, in Pytorch☆88Updated 2 years ago
- ☆203Updated last year
- Implementation of Recurrent Memory Transformer, Neurips 2022 paper, in Pytorch☆412Updated 6 months ago
- [NeurIPS 2022] Official PyTorch implementation of Optimizing Relevance Maps of Vision Transformers Improves Robustness. This code allows …☆131Updated 2 years ago
- A framework for merging models solving different tasks with different initializations into one multi-task model without any additional tr…☆303Updated last year
- ☆208Updated 2 years ago
- Implementation of Mega, the Single-head Attention with Multi-headed EMA architecture that currently holds SOTA on Long Range Arena☆204Updated last year
- Implementation of a memory efficient multi-head attention as proposed in the paper, "Self-attention Does Not Need O(n²) Memory"☆379Updated 2 years ago
- Implementation of the conditionally routed attention in the CoLT5 architecture, in Pytorch☆228Updated 10 months ago
- understanding model mistakes with human annotations☆106Updated 2 years ago
- Just some miscellaneous utility functions / decorators / modules related to Pytorch and Accelerate to help speed up implementation of new…☆124Updated last year
- Implementation of MaMMUT, a simple vision-encoder text-decoder architecture for multimodal tasks from Google, in Pytorch☆103Updated last year
- Code release for "Improved baselines for vision-language pre-training"☆60Updated last year
- Implementation of Block Recurrent Transformer - Pytorch☆220Updated 11 months ago
- ☆182Updated 10 months ago
- [ICML 2023] Official PyTorch implementation of Global Context Vision Transformers☆439Updated last year
- Effortless plugin and play Optimizer to cut model training costs by 50%. New optimizer that is 2x faster than Adam on LLMs.☆379Updated last year
- ☆51Updated last year
- Official code for "TOAST: Transfer Learning via Attention Steering"☆189Updated last year
- Optimized library for large-scale extraction of frames and audio from video.☆204Updated last year
- Implementation of Zorro, Masked Multimodal Transformer, in Pytorch☆97Updated last year
- ☆133Updated last year
- Official PyTorch implementation of the paper: "Solving ImageNet: a Unified Scheme for Training any Backbone to Top Results" (2022)☆193Updated 2 years ago