trainer_state.json

{
  "best_metric": 0.5402666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0381/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 3.1453,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2296,
      "eval_loss": 2.7903218269348145,
      "eval_runtime": 5.8208,
      "eval_samples_per_second": 644.239,
      "eval_steps_per_second": 10.136,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 11.697076797485352,
      "learning_rate": 0.0005,
      "loss": 2.5127,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.34053333333333335,
      "eval_loss": 2.330362558364868,
      "eval_runtime": 6.1056,
      "eval_samples_per_second": 614.191,
      "eval_steps_per_second": 9.663,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 2.1282,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3968,
      "eval_loss": 2.1446287631988525,
      "eval_runtime": 5.7657,
      "eval_samples_per_second": 650.397,
      "eval_steps_per_second": 10.233,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 1.8235,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4528,
      "eval_loss": 1.934847354888916,
      "eval_runtime": 6.4506,
      "eval_samples_per_second": 581.342,
      "eval_steps_per_second": 9.146,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 11.345407485961914,
      "learning_rate": 0.0005,
      "loss": 1.551,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.4922666666666667,
      "eval_loss": 1.8059167861938477,
      "eval_runtime": 5.8007,
      "eval_samples_per_second": 646.472,
      "eval_steps_per_second": 10.171,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 11.446589469909668,
      "learning_rate": 0.0005,
      "loss": 1.2733,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.4952,
      "eval_loss": 1.7578359842300415,
      "eval_runtime": 5.9383,
      "eval_samples_per_second": 631.496,
      "eval_steps_per_second": 9.936,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 23.35776710510254,
      "learning_rate": 0.0005,
      "loss": 1.0169,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.512,
      "eval_loss": 1.8086161613464355,
      "eval_runtime": 5.6344,
      "eval_samples_per_second": 665.554,
      "eval_steps_per_second": 10.471,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.4003744125366211,
      "learning_rate": 0.0005,
      "loss": 0.7565,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5402666666666667,
      "eval_loss": 1.7901859283447266,
      "eval_runtime": 6.7878,
      "eval_samples_per_second": 552.46,
      "eval_steps_per_second": 8.692,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}