trainer_state.json

{
  "best_metric": 0.25,
  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-12",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 12,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 18.849651336669922,
      "learning_rate": 1e-05,
      "loss": 0.8867,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.25,
      "eval_loss": 0.820176899433136,
      "eval_runtime": 19.5446,
      "eval_samples_per_second": 0.614,
      "eval_steps_per_second": 0.614,
      "step": 12
    },
    {
      "epoch": 1.0,
      "step": 12,
      "total_flos": 1.4952775073660928e+16,
      "train_loss": 0.8596544563770294,
      "train_runtime": 49.7949,
      "train_samples_per_second": 0.241,
      "train_steps_per_second": 0.241
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.25,
      "eval_loss": 0.820176899433136,
      "eval_runtime": 19.6934,
      "eval_samples_per_second": 0.609,
      "eval_steps_per_second": 0.609,
      "step": 12
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.25,
      "eval_loss": 0.820176899433136,
      "eval_runtime": 19.472,
      "eval_samples_per_second": 0.616,
      "eval_steps_per_second": 0.616,
      "step": 12
    }
  ],
  "logging_steps": 10,
  "max_steps": 12,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "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": 1.4952775073660928e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}