def training_loop()

in stylegan2_ada_pytorch/training/training_loop.py [0:0]

• 40 lines of code
• 17 McCabe index (conditional complexity)

def training_loop(
    exp_name="default_name",
    run_dir=".",  # Output directory.
    temp_dir=".",  # Temporary directory.
    training_set_kwargs={},  # Options for training set.
    data_loader_kwargs={},  # Options for torch.utils.data.DataLoader.
    G_kwargs={},  # Options for generator network.
    D_kwargs={},  # Options for discriminator network.
    G_opt_kwargs={},  # Options for generator optimizer.
    D_opt_kwargs={},  # Options for discriminator optimizer.
    augment_kwargs=None,  # Options for augmentation pipeline. None = disable.
    loss_kwargs={},  # Options for loss function.
    class_cond=False,  # Condition on class labels.
    instance_cond=False,  # Condition on instance features.
    metrics=[],  # Metrics to evaluate during training.
    random_seed=0,  # Global random seed.
    num_gpus=1,  # Number of GPUs participating in the training.
    slurm=False,  # Launching the experiment in SLURM.
    rank=0,  # Rank of the current process in [0, num_gpus[.
    local_rank=0,  # Local rank of the current process inside each node [0, num_gpus_per_node]
    batch_size=4,  # Total batch size for one training iteration. Can be larger than batch_gpu * num_gpus.
    batch_gpu=4,  # Number of samples processed at a time by one GPU.
    ema_kimg=10,  # Half-life of the exponential moving average (EMA) of generator weights.
    ema_rampup=None,  # EMA ramp-up coefficient.
    G_reg_interval=4,  # How often to perform regularization for G? None = disable lazy regularization.
    D_reg_interval=16,  # How often to perform regularization for D? None = disable lazy regularization.
    augment_p=0,  # Initial value of augmentation probability.
    ada_target=None,  # ADA target value. None = fixed p.
    ada_interval=4,  # How often to perform ADA adjustment?
    ada_kimg=500,  # ADA adjustment speed, measured in how many kimg it takes for p to increase/decrease by one unit.
    total_kimg=25000,  # Total length of the training, measured in thousands of real images.
    kimg_per_tick=4,  # Progress snapshot interval.
    image_snapshot_ticks=50,  # How often to save image snapshots? None = disable.
    network_snapshot_ticks=50,  # How often to save network snapshots? None = disable.
    es_patience=100000000,  # Early stopping patience expressed in number of images seen.
    resume_pkl=None,  # Network pickle to resume training from.
    cudnn_benchmark=True,  # Enable torch.backends.cudnn.benchmark?
    allow_tf32=False,  # Enable torch.backends.cuda.matmul.allow_tf32 and torch.backends.cudnn.allow_tf32?
    abort_fn=None,  # Callback function for determining whether to abort training. Must return consistent results across ranks.
    progress_fn=None,  # Callback function for updating training progress. Called for all ranks.