trainer_state.json

{
  "best_metric": 1.0,
  "best_model_checkpoint": "beit-base-patch16-224-pt22k-ft22k-finetuned-FER2013CKPlus/checkpoint-135",
  "epoch": 4.972972972972973,
  "global_step": 135,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.36,
      "learning_rate": 3.571428571428572e-05,
      "loss": 2.0366,
      "step": 10
    },
    {
      "epoch": 0.72,
      "learning_rate": 4.75206611570248e-05,
      "loss": 0.9918,
      "step": 20
    },
    {
      "epoch": 0.97,
      "eval_accuracy": 0.8984771573604061,
      "eval_loss": 0.252826988697052,
      "eval_runtime": 1.8877,
      "eval_samples_per_second": 104.359,
      "eval_steps_per_second": 6.887,
      "step": 27
    },
    {
      "epoch": 1.11,
      "learning_rate": 4.338842975206612e-05,
      "loss": 0.5841,
      "step": 30
    },
    {
      "epoch": 1.47,
      "learning_rate": 3.925619834710744e-05,
      "loss": 0.4639,
      "step": 40
    },
    {
      "epoch": 1.83,
      "learning_rate": 3.512396694214876e-05,
      "loss": 0.3355,
      "step": 50
    },
    {
      "epoch": 1.97,
      "eval_accuracy": 0.9796954314720813,
      "eval_loss": 0.07027172297239304,
      "eval_runtime": 1.9529,
      "eval_samples_per_second": 100.875,
      "eval_steps_per_second": 6.657,
      "step": 54
    },
    {
      "epoch": 2.22,
      "learning_rate": 3.099173553719008e-05,
      "loss": 0.3101,
      "step": 60
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.6859504132231405e-05,
      "loss": 0.2213,
      "step": 70
    },
    {
      "epoch": 2.94,
      "learning_rate": 2.272727272727273e-05,
      "loss": 0.2484,
      "step": 80
    },
    {
      "epoch": 2.97,
      "eval_accuracy": 0.9847715736040609,
      "eval_loss": 0.023239081725478172,
      "eval_runtime": 1.8186,
      "eval_samples_per_second": 108.323,
      "eval_steps_per_second": 7.148,
      "step": 81
    },
    {
      "epoch": 3.32,
      "learning_rate": 1.859504132231405e-05,
      "loss": 0.2258,
      "step": 90
    },
    {
      "epoch": 3.68,
      "learning_rate": 1.4462809917355372e-05,
      "loss": 0.1971,
      "step": 100
    },
    {
      "epoch": 3.97,
      "eval_accuracy": 0.9847715736040609,
      "eval_loss": 0.01969192549586296,
      "eval_runtime": 1.9453,
      "eval_samples_per_second": 101.271,
      "eval_steps_per_second": 6.683,
      "step": 108
    },
    {
      "epoch": 4.07,
      "learning_rate": 1.0330578512396695e-05,
      "loss": 0.1714,
      "step": 110
    },
    {
      "epoch": 4.43,
      "learning_rate": 6.198347107438017e-06,
      "loss": 0.1553,
      "step": 120
    },
    {
      "epoch": 4.79,
      "learning_rate": 2.066115702479339e-06,
      "loss": 0.1731,
      "step": 130
    },
    {
      "epoch": 4.97,
      "eval_accuracy": 1.0,
      "eval_loss": 0.008882634341716766,
      "eval_runtime": 1.8085,
      "eval_samples_per_second": 108.928,
      "eval_steps_per_second": 7.188,
      "step": 135
    },
    {
      "epoch": 4.97,
      "step": 135,
      "total_flos": 6.807390555083489e+17,
      "train_loss": 0.45934949583477447,
      "train_runtime": 338.9358,
      "train_samples_per_second": 26.037,
      "train_steps_per_second": 0.398
    }
  ],
  "max_steps": 135,
  "num_train_epochs": 5,
  "total_flos": 6.807390555083489e+17,
  "trial_name": null,
  "trial_params": null
}