train404

Attributes

description

Classes

Objective

Functions

setup(rank, world_size, mode)
load_dataset_and_sampler(conf, world_size, rank, is_train)
load_model_states_and_optimizer(conf, model, device)
model_and_memory_summary(conf)
main(rank, world_size, conf[, trial])

Module Contents

train404.setup(rank, world_size, mode)

train404.load_dataset_and_sampler(conf, world_size, rank, is_train, seed=42)

train404.load_model_states_and_optimizer(conf, model, device)

train404.model_and_memory_summary(conf)

train404.main(rank, world_size, conf, trial=False)

class train404.Objective(config, metric='val_loss', device='cpu')

Bases: echo.src.base_objective.BaseObjective

train(trial, conf)

train404.description = 'Train a segmengation model on a hologram data set'