trainer_state.json

{
  "best_metric": 1.0,
  "best_model_checkpoint": "resnet-18/checkpoint-22",
  "epoch": 9.090909090909092,
  "eval_steps": 500,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9090909090909091,
      "eval_accuracy": 0.4155844155844156,
      "eval_loss": 1.0317904949188232,
      "eval_runtime": 0.3039,
      "eval_samples_per_second": 253.336,
      "eval_steps_per_second": 9.87,
      "step": 5
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 2.841965913772583,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 1.0893,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6363636363636364,
      "eval_loss": 0.9519639611244202,
      "eval_runtime": 0.2748,
      "eval_samples_per_second": 280.217,
      "eval_steps_per_second": 10.918,
      "step": 11
    },
    {
      "epoch": 2.909090909090909,
      "eval_accuracy": 0.8441558441558441,
      "eval_loss": 0.9016602039337158,
      "eval_runtime": 0.3011,
      "eval_samples_per_second": 255.709,
      "eval_steps_per_second": 9.963,
      "step": 16
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 3.1325621604919434,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.9912,
      "step": 20
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.8444482088088989,
      "eval_runtime": 0.2042,
      "eval_samples_per_second": 377.122,
      "eval_steps_per_second": 14.693,
      "step": 22
    },
    {
      "epoch": 4.909090909090909,
      "eval_accuracy": 1.0,
      "eval_loss": 0.802738606929779,
      "eval_runtime": 0.221,
      "eval_samples_per_second": 348.461,
      "eval_steps_per_second": 13.576,
      "step": 27
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 4.148845195770264,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.9248,
      "step": 30
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.763073205947876,
      "eval_runtime": 0.2945,
      "eval_samples_per_second": 261.469,
      "eval_steps_per_second": 10.187,
      "step": 33
    },
    {
      "epoch": 6.909090909090909,
      "eval_accuracy": 1.0,
      "eval_loss": 0.736884593963623,
      "eval_runtime": 0.2212,
      "eval_samples_per_second": 348.098,
      "eval_steps_per_second": 13.562,
      "step": 38
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 3.343644142150879,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.8716,
      "step": 40
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.7155924439430237,
      "eval_runtime": 0.261,
      "eval_samples_per_second": 295.058,
      "eval_steps_per_second": 11.496,
      "step": 44
    },
    {
      "epoch": 8.909090909090908,
      "eval_accuracy": 1.0,
      "eval_loss": 0.7137082815170288,
      "eval_runtime": 0.2787,
      "eval_samples_per_second": 276.309,
      "eval_steps_per_second": 10.765,
      "step": 49
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 3.096590757369995,
      "learning_rate": 0.0,
      "loss": 0.8517,
      "step": 50
    },
    {
      "epoch": 9.090909090909092,
      "eval_accuracy": 1.0,
      "eval_loss": 0.7116669416427612,
      "eval_runtime": 0.4991,
      "eval_samples_per_second": 154.284,
      "eval_steps_per_second": 6.011,
      "step": 50
    },
    {
      "epoch": 9.090909090909092,
      "step": 50,
      "total_flos": 6.343101899766374e+16,
      "train_loss": 0.9457284736633301,
      "train_runtime": 45.6784,
      "train_samples_per_second": 151.275,
      "train_steps_per_second": 1.095
    }
  ],
  "logging_steps": 10,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.343101899766374e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}