trainer_state.json

{
  "best_metric": 0.07974568009376526,
  "best_model_checkpoint": "vit-large-patch16-224-dungeon-geo-morphs-011/checkpoint-20",
  "epoch": 23.285714285714285,
  "eval_steps": 10,
  "global_step": 35,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 6.571428571428571,
      "grad_norm": 8.232824325561523,
      "learning_rate": 1.6774193548387098e-05,
      "loss": 0.6594,
      "step": 10
    },
    {
      "epoch": 6.571428571428571,
      "eval_accuracy": 0.9444444444444444,
      "eval_loss": 0.17625682055950165,
      "eval_runtime": 0.5116,
      "eval_samples_per_second": 70.366,
      "eval_steps_per_second": 9.773,
      "step": 10
    },
    {
      "epoch": 13.285714285714286,
      "grad_norm": 0.18349935114383698,
      "learning_rate": 1.0322580645161291e-05,
      "loss": 0.0216,
      "step": 20
    },
    {
      "epoch": 13.285714285714286,
      "eval_accuracy": 0.9444444444444444,
      "eval_loss": 0.07974568009376526,
      "eval_runtime": 0.5091,
      "eval_samples_per_second": 70.708,
      "eval_steps_per_second": 9.821,
      "step": 20
    },
    {
      "epoch": 19.857142857142858,
      "grad_norm": 0.057847198098897934,
      "learning_rate": 3.870967741935484e-06,
      "loss": 0.001,
      "step": 30
    },
    {
      "epoch": 19.857142857142858,
      "eval_accuracy": 0.9444444444444444,
      "eval_loss": 0.1036582887172699,
      "eval_runtime": 0.5527,
      "eval_samples_per_second": 65.136,
      "eval_steps_per_second": 9.047,
      "step": 30
    },
    {
      "epoch": 23.285714285714285,
      "step": 35,
      "total_flos": 3.068073182851891e+17,
      "train_loss": 0.1949317154075418,
      "train_runtime": 179.0363,
      "train_samples_per_second": 9.775,
      "train_steps_per_second": 0.195
    }
  ],
  "logging_steps": 10,
  "max_steps": 35,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 35,
  "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": 3.068073182851891e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}