trainer_state.json

{
  "best_metric": 0.8243243243243243,
  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-114",
  "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": 2.3253,
      "step": 10
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.81203007518797e-05,
      "loss": 2.2018,
      "step": 20
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.43609022556391e-05,
      "loss": 2.1321,
      "step": 30
    },
    {
      "epoch": 0.26,
      "eval_accuracy": 0.5,
      "eval_loss": 1.7063186168670654,
      "eval_runtime": 6.3814,
      "eval_samples_per_second": 11.596,
      "eval_steps_per_second": 1.567,
      "step": 38
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.0601503759398494e-05,
      "loss": 1.8954,
      "step": 40
    },
    {
      "epoch": 1.08,
      "learning_rate": 3.6842105263157895e-05,
      "loss": 1.4623,
      "step": 50
    },
    {
      "epoch": 1.15,
      "learning_rate": 3.3082706766917295e-05,
      "loss": 1.1013,
      "step": 60
    },
    {
      "epoch": 1.22,
      "learning_rate": 2.9323308270676693e-05,
      "loss": 0.7513,
      "step": 70
    },
    {
      "epoch": 1.26,
      "eval_accuracy": 0.6756756756756757,
      "eval_loss": 0.8875435590744019,
      "eval_runtime": 6.6832,
      "eval_samples_per_second": 11.073,
      "eval_steps_per_second": 1.496,
      "step": 76
    },
    {
      "epoch": 2.03,
      "learning_rate": 2.556390977443609e-05,
      "loss": 0.7555,
      "step": 80
    },
    {
      "epoch": 2.09,
      "learning_rate": 2.1804511278195487e-05,
      "loss": 0.531,
      "step": 90
    },
    {
      "epoch": 2.16,
      "learning_rate": 1.8045112781954888e-05,
      "loss": 0.5464,
      "step": 100
    },
    {
      "epoch": 2.23,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.3882,
      "step": 110
    },
    {
      "epoch": 2.26,
      "eval_accuracy": 0.8243243243243243,
      "eval_loss": 0.46954473853111267,
      "eval_runtime": 6.7034,
      "eval_samples_per_second": 11.039,
      "eval_steps_per_second": 1.492,
      "step": 114
    },
    {
      "epoch": 3.04,
      "learning_rate": 1.0526315789473684e-05,
      "loss": 0.271,
      "step": 120
    },
    {
      "epoch": 3.11,
      "learning_rate": 6.766917293233083e-06,
      "loss": 0.2625,
      "step": 130
    },
    {
      "epoch": 3.18,
      "learning_rate": 3.007518796992481e-06,
      "loss": 0.2671,
      "step": 140
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.8243243243243243,
      "eval_loss": 0.3919770121574402,
      "eval_runtime": 6.5967,
      "eval_samples_per_second": 11.218,
      "eval_steps_per_second": 1.516,
      "step": 148
    },
    {
      "epoch": 3.23,
      "step": 148,
      "total_flos": 1.460491890402263e+18,
      "train_loss": 1.0175014428190283,
      "train_runtime": 239.3713,
      "train_samples_per_second": 4.946,
      "train_steps_per_second": 0.618
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.8698224852071006,
      "eval_loss": 0.4558912515640259,
      "eval_runtime": 15.206,
      "eval_samples_per_second": 11.114,
      "eval_steps_per_second": 1.447,
      "step": 148
    },
    {
      "epoch": 3.23,
      "eval_accuracy": 0.8698224852071006,
      "eval_loss": 0.4558912515640259,
      "eval_runtime": 14.3413,
      "eval_samples_per_second": 11.784,
      "eval_steps_per_second": 1.534,
      "step": 148
    }
  ],
  "max_steps": 148,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 1.460491890402263e+18,
  "trial_name": null,
  "trial_params": null
}