
Let an agent drive ML experiments on Polyaxon
Host an agent controller on Polyaxon to launch training and evaluation jobs, inspect results, and choose the next experiment.
The latest guides, tutorials, and product updates.

Host an agent controller on Polyaxon to launch training and evaluation jobs, inspect results, and choose the next experiment.

Compare six experiments at the same epoch, zoom across metric charts together, and configure tooltips with multiple live metric columns and run parameters.

Save a Polyaxon comparison's query, columns, sorting and charts, reopen it for later runs, and share reusable searches and dashboards with your team.

A design note on autonomous experimentation: fixed protocols, selection bias, independent evaluation, and durable research campaigns in Polyaxon.

Tune a PyTorch model's depth, width, dropout, optimizer, and learning rate with native TPE, GPU trials, and validation accuracy in Polyaxon.

Use completed trial results to guide hyperparameter choices with native TPE, run GPU training trials, and compare validation loss in Polyaxon.

Compare old and new application outputs under both evaluator versions, inspect changed decisions, and retain the four comparisons in Polyaxon.

Use an ordered Polyaxon mapping for a curated list of experiments, control concurrency, and distinguish scheduling order from dependencies and completion order.

Use Polyaxon matrices to choose parameter scales, separate meaningful combinations, and control the size of a hyperparameter sweep.

Confirm hyperparameter finalists across matched training seeds, compare variation and paired differences, and reserve compute for a defensible final choice.

Archive a Polyaxon sweep or DAG through its parent run, understand what happens to unfinished children, and restore retained records from Python.

Calibrate early stopping against full training runs, compare equivalent training progress, and track pruning decisions with Polyaxon before enforcing them.