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": 2.4013,
      "step": 10
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.81203007518797e-05,
      "loss": 2.237,
      "step": 20
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.43609022556391e-05,
      "loss": 2.1353,
      "step": 30
    },
    {
      "epoch": 0.26,
      "eval_accuracy": 0.5571428571428572,
      "eval_loss": 1.868154525756836,
      "eval_runtime": 6.7025,
      "eval_samples_per_second": 10.444,
      "eval_steps_per_second": 1.343,
      "step": 38
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.0601503759398494e-05,
      "loss": 1.9805,
      "step": 40
    },
    {
      "epoch": 1.08,
      "learning_rate": 3.6842105263157895e-05,
      "loss": 1.5792,
      "step": 50
    },
    {
      "epoch": 1.15,
      "learning_rate": 3.3082706766917295e-05,
      "loss": 1.2427,
      "step": 60
    },
    {
      "epoch": 1.22,
      "learning_rate": 2.9323308270676693e-05,
      "loss": 0.8568,
      "step": 70
    },
    {
      "epoch": 1.26,
      "eval_accuracy": 0.8,
      "eval_loss": 0.8215473890304565,
      "eval_runtime": 6.87,
      "eval_samples_per_second": 10.189,
      "eval_steps_per_second": 1.31,
      "step": 76
    },
    {
      "epoch": 2.03,
      "learning_rate": 2.556390977443609e-05,
      "loss": 0.841,
      "step": 80
    },
    {
      "epoch": 2.09,
      "learning_rate": 2.1804511278195487e-05,
      "loss": 0.5752,
      "step": 90
    },
    {
      "epoch": 2.16,
      "learning_rate": 1.8045112781954888e-05,
      "loss": 0.5488,
      "step": 100
    },
    {
      "epoch": 2.23,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.4267,
      "step": 110
    },
    {
      "epoch": 2.26,
      "eval_accuracy": 0.8571428571428571,
      "eval_loss": 0.47489041090011597,
      "eval_runtime": 6.8526,
      "eval_samples_per_second": 10.215,
      "eval_steps_per_second": 1.313,
      "step": 114
    },
    {
      "epoch": 3.04,
      "learning_rate": 1.0526315789473684e-05,
      "loss": 0.3191,
      "step": 120
    },
    {
      "epoch": 3.11,
      "learning_rate": 6.766917293233083e-06,
      "loss": 0.2261,
      "step": 130
    },
    {
      "epoch": 3.18,
      "learning_rate": 3.007518796992481e-06,
      "loss": 0.2545,
      "step": 140
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.9142857142857143,
      "eval_loss": 0.317582905292511,
      "eval_runtime": 6.8221,
      "eval_samples_per_second": 10.261,
      "eval_steps_per_second": 1.319,
      "step": 148
    },
    {
      "epoch": 3.23,
      "step": 148,
      "total_flos": 1.460491890402263e+18,
      "train_loss": 1.0695195343043353,
      "train_runtime": 279.5402,
      "train_samples_per_second": 4.236,
      "train_steps_per_second": 0.529
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.8580645161290322,
      "eval_loss": 0.4635992646217346,
      "eval_runtime": 15.8223,
      "eval_samples_per_second": 9.796,
      "eval_steps_per_second": 1.264,
      "step": 148
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.8580645161290322,
      "eval_loss": 0.4635993242263794,
      "eval_runtime": 15.8805,
      "eval_samples_per_second": 9.76,
      "eval_steps_per_second": 1.259,
      "step": 148
    }
  ],
  "max_steps": 148,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 1.460491890402263e+18,
  "trial_name": null,
  "trial_params": null
}