trainer_state.json

{
  "best_metric": 0.05759536102414131,
  "best_model_checkpoint": "vit-large-patch16-224-dungeon-geo-morphs-0-4-28Nov24-001/checkpoint-60",
  "epoch": 24.0,
  "eval_steps": 10,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.0,
      "grad_norm": 25.216020584106445,
      "learning_rate": 1.888888888888889e-05,
      "loss": 1.4212,
      "step": 10
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8428571428571429,
      "eval_loss": 0.8053953051567078,
      "eval_runtime": 7.1329,
      "eval_samples_per_second": 78.51,
      "eval_steps_per_second": 9.814,
      "step": 10
    },
    {
      "epoch": 8.0,
      "grad_norm": 5.78123140335083,
      "learning_rate": 1.5185185185185187e-05,
      "loss": 0.3561,
      "step": 20
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9517857142857142,
      "eval_loss": 0.23307392001152039,
      "eval_runtime": 7.981,
      "eval_samples_per_second": 70.167,
      "eval_steps_per_second": 8.771,
      "step": 20
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.9395315647125244,
      "learning_rate": 1.1481481481481482e-05,
      "loss": 0.0373,
      "step": 30
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9803571428571428,
      "eval_loss": 0.09606237709522247,
      "eval_runtime": 7.7677,
      "eval_samples_per_second": 72.093,
      "eval_steps_per_second": 9.012,
      "step": 30
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.16510935127735138,
      "learning_rate": 7.77777777777778e-06,
      "loss": 0.0042,
      "step": 40
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.9767857142857143,
      "eval_loss": 0.06689313799142838,
      "eval_runtime": 7.1426,
      "eval_samples_per_second": 78.403,
      "eval_steps_per_second": 9.8,
      "step": 40
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.07187718898057938,
      "learning_rate": 4.074074074074074e-06,
      "loss": 0.0013,
      "step": 50
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.975,
      "eval_loss": 0.058019500225782394,
      "eval_runtime": 7.9195,
      "eval_samples_per_second": 70.711,
      "eval_steps_per_second": 8.839,
      "step": 50
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.06839924305677414,
      "learning_rate": 3.7037037037037036e-07,
      "loss": 0.0008,
      "step": 60
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.975,
      "eval_loss": 0.05759536102414131,
      "eval_runtime": 7.7029,
      "eval_samples_per_second": 72.7,
      "eval_steps_per_second": 9.087,
      "step": 60
    },
    {
      "epoch": 24.0,
      "step": 60,
      "total_flos": 5.259589576438579e+17,
      "train_loss": 0.3034758618722359,
      "train_runtime": 356.5119,
      "train_samples_per_second": 6.732,
      "train_steps_per_second": 0.168
    }
  ],
  "logging_steps": 10,
  "max_steps": 60,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.259589576438579e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}