Build and train PyTorch models with nn.Module, DDP, FSDP, mixed precision, and serialization to ONNX.
npx skills add https://github.com/tylertitsworth/skills --skill pytorch-tylertitsworth
Please help me install this Agent Skill. Skill: pytorch Source: https://github.com/tylertitsworth/skills/tree/main/pytorch Command: npx skills add https://github.com/tylertitsworth/skills --skill pytorch-tylertitsworth
PyTorch is the Python deep learning framework that unifies model construction, training loops, and optimization workflows in a way that accelerates experimentation and scalable execution on CPU or GPU.
Train a simple neural network on a small dataset to validate your environment.
High-intent search queries and answers about installing and using this skill.
PyTorch model construction uses nn.Module to define layers and forward passes, leveraging dynamic graphs and autograd for flexible research prototyping and production training loops.
Distributed training in PyTorch scales deep learning across CPU and GPU clusters using Distributed Data Parallel (DDP) and Fully Sharded Data Parallel (FSDP) to efficiently partition computation and memory.
Mixed precision training in PyTorch uses autocast to automatically select lower precision operations and GradScaler to prevent gradient underflow, optimizing performance and memory usage during model training.
Yes, PyTorch supports model serialization and deployment through state_dict for saving weights, TorchScript for portable graph representation, and ONNX for cross-platform framework interoperability and inference.
PyTorch supports performance optimization through torch.compile, which dynamically compiles model code to reduce overhead and accelerate execution across research prototyping and scalable production environments.
hung-phanBuild and train PyTorch deep learning models with custom training loops.
tondevrelExport PyTorch models to TorchScript and ONNX for deployment.
ksopylaGuide PyTorch neural network design with tensor shape sketching and memory analysis.
davpatel605-beepConvert PyTorch training loops into Lightning Trainer workflows with distributed support.
