trainer_state.json

{
  "best_metric": 0.4889937106918239,
  "best_model_checkpoint": "resnet-50-finetuned-student_kaggle/checkpoint-11",
  "epoch": 4.680851063829787,
  "eval_steps": 500,
  "global_step": 55,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.851063829787234,
      "grad_norm": NaN,
      "learning_rate": 4.591836734693878e-05,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.9361702127659575,
      "eval_accuracy": 0.4889937106918239,
      "eval_loss": NaN,
      "eval_runtime": 8.3443,
      "eval_samples_per_second": 76.22,
      "eval_steps_per_second": 2.397,
      "step": 11
    },
    {
      "epoch": 1.702127659574468,
      "grad_norm": NaN,
      "learning_rate": 3.571428571428572e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 1.9574468085106385,
      "eval_accuracy": 0.4889937106918239,
      "eval_loss": NaN,
      "eval_runtime": 7.5463,
      "eval_samples_per_second": 84.28,
      "eval_steps_per_second": 2.65,
      "step": 23
    },
    {
      "epoch": 2.5531914893617023,
      "grad_norm": NaN,
      "learning_rate": 2.5510204081632654e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 2.978723404255319,
      "eval_accuracy": 0.4889937106918239,
      "eval_loss": NaN,
      "eval_runtime": 8.3005,
      "eval_samples_per_second": 76.621,
      "eval_steps_per_second": 2.409,
      "step": 35
    },
    {
      "epoch": 3.404255319148936,
      "grad_norm": NaN,
      "learning_rate": 1.5306122448979594e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4889937106918239,
      "eval_loss": NaN,
      "eval_runtime": 14.6795,
      "eval_samples_per_second": 43.326,
      "eval_steps_per_second": 1.362,
      "step": 47
    },
    {
      "epoch": 4.25531914893617,
      "grad_norm": NaN,
      "learning_rate": 5.102040816326531e-06,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 4.680851063829787,
      "eval_accuracy": 0.4889937106918239,
      "eval_loss": NaN,
      "eval_runtime": 7.5742,
      "eval_samples_per_second": 83.969,
      "eval_steps_per_second": 2.641,
      "step": 55
    },
    {
      "epoch": 4.680851063829787,
      "step": 55,
      "total_flos": 1.477984078577664e+17,
      "train_loss": 0.0,
      "train_runtime": 167.7096,
      "train_samples_per_second": 44.243,
      "train_steps_per_second": 0.328
    }
  ],
  "logging_steps": 10,
  "max_steps": 55,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.477984078577664e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}