{"repo":"traceopt-ai/traceml","free":true,"listed":false,"github":"https://github.com/traceopt-ai/traceml","clone":"git clone https://github.com/traceopt-ai/traceml.git","description":"Open-source performance diagnostics for PyTorch training runs.","language":"Python","stars":183,"topics":["pytorch","ddp","distributed-training","gpu","hugging-face","pytorch-lightning","deep-learning","machine-learning","mlops","ray"],"license":"Apache-2.0","category":"machine-learning","readme_excerpt":"TraceML Find out why your PyTorch training is slow—before it wastes GPU hours. Quickstart • Try in Colab • Integrations • Documentation • Discord TraceML is an open-source tool that explains why PyTorch training is slow. It analyzes the full training run and gives you: - A diagnosis: data loading, GPU compute, waiting, memory growth, or a slow distributed worker. - The evidence: timing, CPU/GPU usage, memory, and per-worker measurements. - The next step: what part of your training setup to investigate first. Here is an example where TraceML finds that a slow DataLoader is leaving the GPU waiting: Running distributed training? See a rank-straggler diagnosis Quickstart 1. Install Or, in a project managed by uv: 2. Instrument the training step 3. Run Summary mode is the default. TraceML prints the final diagnosis and writes: See the full quickstart for source-only examples, Docker (both require a repository checkout), Colab, direct python / torchrun launches, HTML reports, and advanced options. What TraceML Diagnoses Diagnosis Where to investigate --- --- Input-bound DataLoader workers, transforms, tokenization, collation, or storage Compute-bound Model compute, mixed precision, batch size, or deeper profiling Residual-heavy Work outside traced phases, CPU stalls, logging, checkpointing, validation, or unobserved transfers Rank straggler Rank-local input, data imbalance, node variance, or networking Memory creep Retained tensors, logging references, or cached activations Run reg","default_branch":null,"files":null,"tree":[],"storefront":"/r/traceopt-ai","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/traceopt-ai/traceml/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}