Estimate Runtime

Run a new run script and compare its runtime to SGD.

usage: deepobs_estimate_runtime.py [-h] [--test_problem TEST_PROBLEM]
                                   [--data_dir DATA_DIR] [--bs BS] [--lr LR]
                                   [-N NUM_EPOCHS] [--num_runs NUM_RUNS]
                                   [--saveto SAVETO]
                                   [--optimizer_arguments OPTIMIZER_ARGUMENTS]
                                   run_script

Positional Arguments

run_script

path to the new run_script.

Named Arguments

--test_problem	Name of the test problem to run both scripts. Default: "mnist.mnist_mlp"
--data_dir	Path to the base data dir. If not set, deepobs uses its default. Default: "data_deepobs"
--bs, --batch_size
	The batch size (positive integer). Defaults to 128. Default: 128
--lr, --learning_rate
	The learning rate of SGD, defaults to 1e-5. Default: 1e-05
-N, --num_epochs
	Total number of training steps. Default: 3
--num_runs	Total number of runs for each optimizer. Default: 5
--saveto	Folder for saving a txt files with a summary.
--optimizer_arguments
	Additional arguments for the new optimizer