trainer_state.json

{
  "best_metric": 0.9,
  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-148",
  "epoch": 3.22972972972973,
  "global_step": 148,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 2.3721,
      "step": 10
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.81203007518797e-05,
      "loss": 2.2408,
      "step": 20
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.43609022556391e-05,
      "loss": 2.0323,
      "step": 30
    },
    {
      "epoch": 0.26,
      "eval_accuracy": 0.5571428571428572,
      "eval_loss": 1.6401814222335815,
      "eval_runtime": 132.7593,
      "eval_samples_per_second": 0.527,
      "eval_steps_per_second": 0.068,
      "step": 38
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.0601503759398494e-05,
      "loss": 1.873,
      "step": 40
    },
    {
      "epoch": 1.08,
      "learning_rate": 3.6842105263157895e-05,
      "loss": 1.4581,
      "step": 50
    },
    {
      "epoch": 1.15,
      "learning_rate": 3.3082706766917295e-05,
      "loss": 1.1944,
      "step": 60
    },
    {
      "epoch": 1.22,
      "learning_rate": 2.9323308270676693e-05,
      "loss": 0.8081,
      "step": 70
    },
    {
      "epoch": 1.26,
      "eval_accuracy": 0.6857142857142857,
      "eval_loss": 0.8534234762191772,
      "eval_runtime": 133.5733,
      "eval_samples_per_second": 0.524,
      "eval_steps_per_second": 0.067,
      "step": 76
    },
    {
      "epoch": 2.03,
      "learning_rate": 2.556390977443609e-05,
      "loss": 0.7699,
      "step": 80
    },
    {
      "epoch": 2.09,
      "learning_rate": 2.1804511278195487e-05,
      "loss": 0.5961,
      "step": 90
    },
    {
      "epoch": 2.16,
      "learning_rate": 1.8045112781954888e-05,
      "loss": 0.5411,
      "step": 100
    },
    {
      "epoch": 2.23,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.4099,
      "step": 110
    },
    {
      "epoch": 2.26,
      "eval_accuracy": 0.8428571428571429,
      "eval_loss": 0.5761450529098511,
      "eval_runtime": 131.9517,
      "eval_samples_per_second": 0.53,
      "eval_steps_per_second": 0.068,
      "step": 114
    },
    {
      "epoch": 3.04,
      "learning_rate": 1.0526315789473684e-05,
      "loss": 0.3393,
      "step": 120
    },
    {
      "epoch": 3.11,
      "learning_rate": 6.766917293233083e-06,
      "loss": 0.2812,
      "step": 130
    },
    {
      "epoch": 3.18,
      "learning_rate": 3.007518796992481e-06,
      "loss": 0.255,
      "step": 140
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.9,
      "eval_loss": 0.39665931463241577,
      "eval_runtime": 127.4255,
      "eval_samples_per_second": 0.549,
      "eval_steps_per_second": 0.071,
      "step": 148
    },
    {
      "epoch": 3.23,
      "step": 148,
      "total_flos": 1.460491890402263e+18,
      "train_loss": 1.0397279697495538,
      "train_runtime": 7461.1258,
      "train_samples_per_second": 0.159,
      "train_steps_per_second": 0.02
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.8580645161290322,
      "eval_loss": 0.5035645365715027,
      "eval_runtime": 298.4366,
      "eval_samples_per_second": 0.519,
      "eval_steps_per_second": 0.067,
      "step": 148
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.8580645161290322,
      "eval_loss": 0.5035644769668579,
      "eval_runtime": 299.9496,
      "eval_samples_per_second": 0.517,
      "eval_steps_per_second": 0.067,
      "step": 148
    }
  ],
  "max_steps": 148,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 1.460491890402263e+18,
  "trial_name": null,
  "trial_params": null
}