Class-Balanced Loss Based on Effective Number of Samples. CVPR 2019
-
Updated
Aug 29, 2021 - Python
Class-Balanced Loss Based on Effective Number of Samples. CVPR 2019
Everything you want to know about Google Cloud TPU
Keras implementation of the renowned publication "DeepFace: Closing the Gap to Human-Level Performance in Face Verification" by Taigman et al. Pre-trained weights on VGGFace2 dataset.
Hardware-grounded model architecture design principles, MoE co-design patterns, low-precision numerics, and parallelism guides for training and serving on Google Cloud TPUs.
Automated provisioner of a Google Cloud TPU environment for training in PyTorch
NLP Code Snippets and Conference related
A collection of CNN models are trained on Cloud TPU by using PyTorch/XLA
Tutorial: mass-parallelized RAI compliance checks with vLLM batch inference on Cloud TPU v5e, using Gemma 4 as an LLM-as-a-Judge
C++20 / OpenXLA PJRT inference engine for Google Cloud TPUs: static HBM arenas, 128/256-row systolic tree speculation, INT8/FP8 kernels, and Gemma 4 serving.
Interactive terminal monitor for Google Cloud TPU
Empirical Roofline, OpenXLA compiler, and cost benchmarks across NVIDIA V100, NVIDIA L4, Cloud TPU v5e, and Cloud TPU v6e Trillium on a campus Slurm cluster (doi:10.5281/zenodo.24224726)
To associate your repository with the cloud-tpu topic, visit your repo's landing page and select "manage topics."