Deeplearning

• FlashAttention - optimizing GPU memory for more scalable transformers Jul 20
• LoRA - low rank adaption explained in three minutes Jan 28

• Understanding the difference between weight decay and L2 regularization Sep 17
• Semantic segmentation with prototype-based consistency regularization Jan 29

• Everything you need to know about stable diffusion Nov 16
• How and why stable diffusion works for text to image generation Aug 27
• Rethinking Depthwise Separable Convolutions in PyTorch Jul 19
• DINO - Emerging properties in self-supervised vision transformers Mar 13
• Rethinking Batch in BatchNorm Feb 28

• Hyperparameter tuning on numerai data with PyTorch Lightning and weights & biases Dec 5