nerfplayer method_configs

method_configs["nerfplayer"] = TrainerConfig(
    method_name="nerfplayer",
    steps_per_eval_batch=1000,
    steps_per_eval_all_images=0,
    steps_per_eval_image=500,
    steps_per_save=10000,
    save_only_latest_checkpoint=False,
    max_num_iterations=30000,
    mixed_precision=True,
    pipeline=VanillaPipelineConfig(
        datamanager=VanillaDataManagerConfig(
            dataparser=BlenderDataParserConfig(),
            train_num_rays_per_batch=4096,
            eval_num_rays_per_batch=2048,
        ),
        model=NerfplayerModelConfig(
            eval_num_rays_per_chunk=1 << 15,
            log2_hashmap_size=17,
            temporal_dim=64,
            depth_weight=1.0,
            depth_sigma=0.01,
            prob_reg_loss_mult=0.1,
            distortion_loss_mult=0.001,
            temporal_tv_weight=1.0,
        ),
    ),
    optimizers={
        "proposal_networks": {
            "optimizer": AdamOptimizerConfig(lr=1e-2, eps=1e-12),
            "scheduler": CosineDecaySchedulerConfig(warm_up_end=512, max_steps=30000, learning_rate_alpha=0),
        },
        "fields": {
            "optimizer": AdamOptimizerConfig(lr=1e-2, eps=1e-12),
            "scheduler": CosineDecaySchedulerConfig(warm_up_end=512, max_steps=30000, learning_rate_alpha=0),
        },
    },
    viewer=ViewerConfig(num_rays_per_chunk=64000),
    vis="viewer",
)