trainer_state.json

{
  "best_metric": 0.8802666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0431/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": 9.621650897886543e-05,
      "loss": 1.5546,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7722666666666667,
      "eval_loss": 0.7884587049484253,
      "eval_runtime": 8.1498,
      "eval_samples_per_second": 460.136,
      "eval_steps_per_second": 7.239,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 8.543862808774192e-05,
      "loss": 0.5859,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.82,
      "eval_loss": 0.5805087685585022,
      "eval_runtime": 7.3643,
      "eval_samples_per_second": 509.212,
      "eval_steps_per_second": 8.012,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.283578872680664,
      "learning_rate": 6.924306924189557e-05,
      "loss": 0.3702,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8338666666666666,
      "eval_loss": 0.5215000510215759,
      "eval_runtime": 7.0997,
      "eval_samples_per_second": 528.194,
      "eval_steps_per_second": 8.31,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.2808640003204346,
      "learning_rate": 5.011792754282769e-05,
      "loss": 0.2383,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8512,
      "eval_loss": 0.48370522260665894,
      "eval_runtime": 7.7519,
      "eval_samples_per_second": 483.75,
      "eval_steps_per_second": 7.611,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.070003867149353,
      "learning_rate": 3.0974832444380136e-05,
      "loss": 0.138,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8693333333333333,
      "eval_loss": 0.4408036172389984,
      "eval_runtime": 7.1252,
      "eval_samples_per_second": 526.3,
      "eval_steps_per_second": 8.28,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.05040293186903,
      "learning_rate": 1.472814664270234e-05,
      "loss": 0.0753,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8725333333333334,
      "eval_loss": 0.44236263632774353,
      "eval_runtime": 6.8906,
      "eval_samples_per_second": 544.218,
      "eval_steps_per_second": 8.562,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.7301777601242065,
      "learning_rate": 3.851280774526222e-06,
      "loss": 0.0405,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8768,
      "eval_loss": 0.4306723177433014,
      "eval_runtime": 7.0554,
      "eval_samples_per_second": 531.509,
      "eval_steps_per_second": 8.362,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.8271176218986511,
      "learning_rate": 1.3906924697382196e-10,
      "loss": 0.0258,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8802666666666666,
      "eval_loss": 0.43250203132629395,
      "eval_runtime": 7.6698,
      "eval_samples_per_second": 488.931,
      "eval_steps_per_second": 7.693,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}