trainer_state.json

{
  "best_metric": 0.9142857142857143,
  "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": 2.3117,
      "step": 10
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.81203007518797e-05,
      "loss": 2.1973,
      "step": 20
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.43609022556391e-05,
      "loss": 2.143,
      "step": 30
    },
    {
      "epoch": 0.26,
      "eval_accuracy": 0.5,
      "eval_loss": 1.8738905191421509,
      "eval_runtime": 6.1995,
      "eval_samples_per_second": 11.291,
      "eval_steps_per_second": 1.452,
      "step": 38
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.0601503759398494e-05,
      "loss": 2.0085,
      "step": 40
    },
    {
      "epoch": 1.08,
      "learning_rate": 3.6842105263157895e-05,
      "loss": 1.6213,
      "step": 50
    },
    {
      "epoch": 1.15,
      "learning_rate": 3.3082706766917295e-05,
      "loss": 1.322,
      "step": 60
    },
    {
      "epoch": 1.22,
      "learning_rate": 2.9323308270676693e-05,
      "loss": 0.9177,
      "step": 70
    },
    {
      "epoch": 1.26,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.9705972075462341,
      "eval_runtime": 6.3319,
      "eval_samples_per_second": 11.055,
      "eval_steps_per_second": 1.421,
      "step": 76
    },
    {
      "epoch": 2.03,
      "learning_rate": 2.556390977443609e-05,
      "loss": 0.9302,
      "step": 80
    },
    {
      "epoch": 2.09,
      "learning_rate": 2.1804511278195487e-05,
      "loss": 0.6676,
      "step": 90
    },
    {
      "epoch": 2.16,
      "learning_rate": 1.8045112781954888e-05,
      "loss": 0.585,
      "step": 100
    },
    {
      "epoch": 2.23,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.4706,
      "step": 110
    },
    {
      "epoch": 2.26,
      "eval_accuracy": 0.8428571428571429,
      "eval_loss": 0.5700908899307251,
      "eval_runtime": 6.2602,
      "eval_samples_per_second": 11.182,
      "eval_steps_per_second": 1.438,
      "step": 114
    },
    {
      "epoch": 3.04,
      "learning_rate": 1.0526315789473684e-05,
      "loss": 0.3946,
      "step": 120
    },
    {
      "epoch": 3.11,
      "learning_rate": 6.766917293233083e-06,
      "loss": 0.2974,
      "step": 130
    },
    {
      "epoch": 3.18,
      "learning_rate": 3.007518796992481e-06,
      "loss": 0.2895,
      "step": 140
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.9142857142857143,
      "eval_loss": 0.41213229298591614,
      "eval_runtime": 6.1617,
      "eval_samples_per_second": 11.36,
      "eval_steps_per_second": 1.461,
      "step": 148
    },
    {
      "epoch": 3.23,
      "step": 148,
      "total_flos": 1.460491890402263e+18,
      "train_loss": 1.1060102565868482,
      "train_runtime": 228.6736,
      "train_samples_per_second": 5.178,
      "train_steps_per_second": 0.647
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.8516129032258064,
      "eval_loss": 0.5003896951675415,
      "eval_runtime": 14.2101,
      "eval_samples_per_second": 10.908,
      "eval_steps_per_second": 1.407,
      "step": 148
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.8516129032258064,
      "eval_loss": 0.5003897547721863,
      "eval_runtime": 13.4341,
      "eval_samples_per_second": 11.538,
      "eval_steps_per_second": 1.489,
      "step": 148
    }
  ],
  "max_steps": 148,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 1.460491890402263e+18,
  "trial_name": null,
  "trial_params": null
}