trainer_state.json

{
  "best_metric": 0.2625,
  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-25",
  "epoch": 3.84,
  "eval_steps": 500,
  "global_step": 96,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4,
      "grad_norm": 9.724103927612305,
      "learning_rate": 5e-05,
      "loss": 1.4798,
      "step": 10
    },
    {
      "epoch": 0.8,
      "grad_norm": 8.079227447509766,
      "learning_rate": 4.418604651162791e-05,
      "loss": 1.4528,
      "step": 20
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2625,
      "eval_loss": 1.4395983219146729,
      "eval_runtime": 5.2554,
      "eval_samples_per_second": 15.222,
      "eval_steps_per_second": 1.903,
      "step": 25
    },
    {
      "epoch": 1.2,
      "grad_norm": 3.919217824935913,
      "learning_rate": 3.837209302325582e-05,
      "loss": 1.4901,
      "step": 30
    },
    {
      "epoch": 1.6,
      "grad_norm": 1.8412970304489136,
      "learning_rate": 3.2558139534883724e-05,
      "loss": 1.4273,
      "step": 40
    },
    {
      "epoch": 2.0,
      "grad_norm": 7.860177993774414,
      "learning_rate": 2.674418604651163e-05,
      "loss": 1.4244,
      "step": 50
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.25,
      "eval_loss": 1.4033204317092896,
      "eval_runtime": 4.221,
      "eval_samples_per_second": 18.953,
      "eval_steps_per_second": 2.369,
      "step": 50
    },
    {
      "epoch": 2.4,
      "grad_norm": 5.914783000946045,
      "learning_rate": 2.0930232558139536e-05,
      "loss": 1.4058,
      "step": 60
    },
    {
      "epoch": 2.8,
      "grad_norm": 3.5916202068328857,
      "learning_rate": 1.5116279069767441e-05,
      "loss": 1.4495,
      "step": 70
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.25,
      "eval_loss": 1.3919236660003662,
      "eval_runtime": 4.5151,
      "eval_samples_per_second": 17.718,
      "eval_steps_per_second": 2.215,
      "step": 75
    },
    {
      "epoch": 3.2,
      "grad_norm": 5.373912811279297,
      "learning_rate": 9.302325581395349e-06,
      "loss": 1.4098,
      "step": 80
    },
    {
      "epoch": 3.6,
      "grad_norm": 5.40514612197876,
      "learning_rate": 3.488372093023256e-06,
      "loss": 1.3912,
      "step": 90
    },
    {
      "epoch": 3.84,
      "eval_accuracy": 0.25,
      "eval_loss": 1.3884985446929932,
      "eval_runtime": 4.4577,
      "eval_samples_per_second": 17.947,
      "eval_steps_per_second": 2.243,
      "step": 96
    },
    {
      "epoch": 3.84,
      "step": 96,
      "total_flos": 9.420416318601953e+17,
      "train_loss": 1.4338702062765758,
      "train_runtime": 108.4417,
      "train_samples_per_second": 7.082,
      "train_steps_per_second": 0.885
    },
    {
      "epoch": 3.84,
      "eval_accuracy": 0.23684210526315788,
      "eval_loss": 1.4163174629211426,
      "eval_runtime": 4.7313,
      "eval_samples_per_second": 16.063,
      "eval_steps_per_second": 2.114,
      "step": 96
    },
    {
      "epoch": 3.84,
      "eval_accuracy": 0.23684210526315788,
      "eval_loss": 1.4163174629211426,
      "eval_runtime": 4.1873,
      "eval_samples_per_second": 18.15,
      "eval_steps_per_second": 2.388,
      "step": 96
    }
  ],
  "logging_steps": 10,
  "max_steps": 96,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9.420416318601953e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}