trainer_state.json

{
  "best_metric": 0.7571428571428571,
  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-300",
  "epoch": 1.5,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.33,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 2.0705,
      "step": 100
    },
    {
      "epoch": 0.5,
      "eval_accuracy": 0.6285714285714286,
      "eval_loss": 1.0643583536148071,
      "eval_runtime": 12.0604,
      "eval_samples_per_second": 5.804,
      "eval_steps_per_second": 2.902,
      "step": 150
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 1.1414,
      "step": 200
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0,
      "loss": 0.555,
      "step": 300
    },
    {
      "epoch": 1.5,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.4973643720149994,
      "eval_runtime": 14.1361,
      "eval_samples_per_second": 4.952,
      "eval_steps_per_second": 2.476,
      "step": 300
    },
    {
      "epoch": 1.5,
      "step": 300,
      "total_flos": 7.47692094062592e+17,
      "train_loss": 1.2556298955281575,
      "train_runtime": 270.1133,
      "train_samples_per_second": 2.221,
      "train_steps_per_second": 1.111
    },
    {
      "epoch": 1.5,
      "eval_accuracy": 0.832258064516129,
      "eval_loss": 0.45837825536727905,
      "eval_runtime": 34.7949,
      "eval_samples_per_second": 4.455,
      "eval_steps_per_second": 2.242,
      "step": 300
    },
    {
      "epoch": 1.5,
      "eval_accuracy": 0.832258064516129,
      "eval_loss": 0.45837831497192383,
      "eval_runtime": 35.474,
      "eval_samples_per_second": 4.369,
      "eval_steps_per_second": 2.199,
      "step": 300
    }
  ],
  "max_steps": 300,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 7.47692094062592e+17,
  "trial_name": null,
  "trial_params": null
}