LabML

What LabML is

LabML is an open-source platform for monitoring deep learning experiments and hardware usage. It can show experiment results, hardware usage, and performance metrics in real time on mobile devices and laptops. Users can record metrics such as training loss, track hyperparameters and Git commits, build custom charts, and compare runs. The platform also supports distributed training across multiple machines. Setup uses pip to install the app and `labml app-server` to start a local server; hosting the experiments server requires MongoDB. Documentation lists sample integrations for FastAI, Keras, and PyTorch Lightning, along with Python API and CLI references. A related collection provides documented PyTorch implementations covering topics such as transformers, diffusion models, reinforcement learning, optimizers, and normalization layers; it installs with `pip install labml-nn`. LabML software is provided under the MIT License, which states it comes without warranty. The site lists [email protected] for inquiries.