config.json

{
  "accelerator_kwargs": {},
  "allow_tf32": true,
  "async_reward_computation": false,
  "dataset_index": 11,
  "exp_name": "o2o",
  "global_step": 0,
  "high_reward": 100.0,
  "huggingface_note": "B50s100x5x2d2Maycode",
  "kl_weight": 0.0001,
  "log_with": "wandb",
  "logdir": "ll",
  "low_reward": 50.0,
  "lr_scheduler": "linear",
  "max_loop": 50,
  "max_workers": 2,
  "mixed_precision": "fp16",
  "negative_prompts": "",
  "num_checkpoint_limit": 10,
  "num_epochs": 5,
  "offpolicy_sample_batch_size": 5,
  "online_multification_number": 1,
  "per_prompt_stat_tracking": false,
  "per_prompt_stat_tracking_buffer_size": 16,
  "per_prompt_stat_tracking_min_count": 16,
  "project_kwargs": {
    "automatic_checkpoint_naming": true,
    "logging_dir": "./outputs/logs",
    "project_dir": "./outputs/",
    "total_limit": 5
  },
  "resolution": 768,
  "resume_from": "",
  "run_name": "",
  "sample_eta": 1.0,
  "sample_guidance_scale": 5.0,
  "sample_num_batches_per_loop": 1,
  "sample_num_steps": 100,
  "save_freq": 20,
  "seed": 1,
  "show_metrics": true,
  "tracker_kwargs": {},
  "tracker_project_name": "Dev",
  "train_adam_beta1": 0.9,
  "train_adam_beta2": 0.999,
  "train_adam_epsilon": 1e-08,
  "train_adam_weight_decay": 0.0001,
  "train_adv_clip_max": 5,
  "train_batch_size": 5,
  "train_cfg": true,
  "train_clip_range": 0.0001,
  "train_gradient_accumulation_steps": 1,
  "train_learning_rate": 0.0003,
  "train_max_grad_norm": 1.0,
  "train_num_inner_loops": 1,
  "train_timestep_fraction": 1.0,
  "train_use_8bit_adam": false,
  "usage_dataset_size": 50,
  "vila_stopping": 0.4
}