Skill UI

Collection of battle-tested PyTorch training recipes covering architecture selection, optimizers, LR scheduling, precision tips, and debugging to train or fine-tune LLMs, vision, diffusion, and biomedical models with high GPU throughput.

A comprehensive skill for managing the entire lifecycle of Volcano Jobs within the KubeSphere environment. It supports the scheduling and execution of complex batch workloads, including deep learning tasks using PyTorch and TensorFlow, MPI, and general resource-intensive jobs. Provides detailed API usage and troubleshooting guidance for job scheduling and resource conflicts.