trainer_state.json

{
  "best_metric": 0.9285714285714286,
  "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": 2.3623,
      "step": 10
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.81203007518797e-05,
      "loss": 2.2155,
      "step": 20
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.43609022556391e-05,
      "loss": 2.1374,
      "step": 30
    },
    {
      "epoch": 0.26,
      "eval_accuracy": 0.5714285714285714,
      "eval_loss": 1.7412834167480469,
      "eval_runtime": 6.1835,
      "eval_samples_per_second": 11.32,
      "eval_steps_per_second": 1.455,
      "step": 38
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.0601503759398494e-05,
      "loss": 1.9183,
      "step": 40
    },
    {
      "epoch": 1.08,
      "learning_rate": 3.6842105263157895e-05,
      "loss": 1.4923,
      "step": 50
    },
    {
      "epoch": 1.15,
      "learning_rate": 3.3082706766917295e-05,
      "loss": 1.2149,
      "step": 60
    },
    {
      "epoch": 1.22,
      "learning_rate": 2.9323308270676693e-05,
      "loss": 0.7949,
      "step": 70
    },
    {
      "epoch": 1.26,
      "eval_accuracy": 0.8,
      "eval_loss": 0.7746667265892029,
      "eval_runtime": 6.3565,
      "eval_samples_per_second": 11.012,
      "eval_steps_per_second": 1.416,
      "step": 76
    },
    {
      "epoch": 2.03,
      "learning_rate": 2.556390977443609e-05,
      "loss": 0.8088,
      "step": 80
    },
    {
      "epoch": 2.09,
      "learning_rate": 2.1804511278195487e-05,
      "loss": 0.535,
      "step": 90
    },
    {
      "epoch": 2.16,
      "learning_rate": 1.8045112781954888e-05,
      "loss": 0.5405,
      "step": 100
    },
    {
      "epoch": 2.23,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.4279,
      "step": 110
    },
    {
      "epoch": 2.26,
      "eval_accuracy": 0.9142857142857143,
      "eval_loss": 0.4053409993648529,
      "eval_runtime": 6.2263,
      "eval_samples_per_second": 11.243,
      "eval_steps_per_second": 1.445,
      "step": 114
    },
    {
      "epoch": 3.04,
      "learning_rate": 1.0526315789473684e-05,
      "loss": 0.3461,
      "step": 120
    },
    {
      "epoch": 3.11,
      "learning_rate": 6.766917293233083e-06,
      "loss": 0.2509,
      "step": 130
    },
    {
      "epoch": 3.18,
      "learning_rate": 3.007518796992481e-06,
      "loss": 0.291,
      "step": 140
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.9285714285714286,
      "eval_loss": 0.3428815007209778,
      "eval_runtime": 6.583,
      "eval_samples_per_second": 10.633,
      "eval_steps_per_second": 1.367,
      "step": 148
    },
    {
      "epoch": 3.23,
      "step": 148,
      "total_flos": 1.460491890402263e+18,
      "train_loss": 1.0492687128685616,
      "train_runtime": 239.2161,
      "train_samples_per_second": 4.949,
      "train_steps_per_second": 0.619
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.864516129032258,
      "eval_loss": 0.3992488980293274,
      "eval_runtime": 14.1288,
      "eval_samples_per_second": 10.971,
      "eval_steps_per_second": 1.416,
      "step": 148
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.864516129032258,
      "eval_loss": 0.3992488980293274,
      "eval_runtime": 13.7986,
      "eval_samples_per_second": 11.233,
      "eval_steps_per_second": 1.449,
      "step": 148
    }
  ],
  "max_steps": 148,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 1.460491890402263e+18,
  "trial_name": null,
  "trial_params": null
}