Skip to content

Learning Rate Finder Guide

Use this tool after completing Workflow Step 3 and before full training (Workflow Step 4). The workflow shows where LR Finder fits; this page highlights how to interpret results and when to tweak options.


Why Run It?

  • Quickly identify a safe learning-rate window for a new dataset/model.
  • Avoid repeated manual sweeps across magnitudes (1e-4 → 1e-2 → ...).
  • Catch obviously unstable configurations before long training runs.

A single command launches the scan:

ml-lr-finder --config configs/my_dataset_config.yaml

The command writes a plot and JSON under runs/lr_finder_<timestamp>/ and prints a suggested LR.


Reading the Plot

  • Steep descent region → best learning-rate range.
  • Flat left side → LR too small, training will be slow.
  • Sharp spike upwards → LR too large, loss diverging (stop uses --diverge_threshold).
  • Pick a value slightly before the curve turns upward; applying ~½ to 1× the suggested LR is usually stable.

Update your config or call ml-train ... --lr <value> accordingly.


Useful Flags

Flag Default Use it when
--start_lr 1e-7 Your model tolerates larger minimum LRs
--end_lr 1 Smaller models require a lower upper bound
--num_iter 100 Increase for smoother curves, decrease for speed
--beta 0.98 Controls loss smoothing; lower for noisier data
--diverge_threshold 4.0 Set lower to stop earlier on sensitive models
--fold current fold Align finder with the fold you will train

Example:

ml-lr-finder --config configs/my_dataset_config.yaml   --start_lr 1e-6 --end_lr 5e-2 --num_iter 150 --diverge_threshold 3.0


Tips & Troubleshooting

  • Suggestion looks too high → use 0.1× the suggested LR or tighten --diverge_threshold.
  • Curve noisy → increase --num_iter or reduce --beta so smoothing reacts faster.
  • Finder crashes with OOM → lower --batch_size temporarily or run on CPU.
  • Different folds behave differently → rerun finder per fold when datasets are imbalanced.

If you already have an LR from a previous run, you can skip the finder and proceed directly to training.