trainer_state.json

{
  "best_metric": 0.9136,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0094/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": 6.820662498474121,
      "learning_rate": 0.0004809698831278217,
      "loss": 0.832,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7901333333333334,
      "eval_loss": 0.7218156456947327,
      "eval_runtime": 14.5041,
      "eval_samples_per_second": 258.547,
      "eval_steps_per_second": 4.068,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0004269850409335979,
      "loss": 0.4763,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8381333333333333,
      "eval_loss": 0.5536264181137085,
      "eval_runtime": 12.4664,
      "eval_samples_per_second": 300.809,
      "eval_steps_per_second": 4.733,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.30234232544898987,
      "learning_rate": 0.0003459431688641196,
      "loss": 0.2986,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8554666666666667,
      "eval_loss": 0.527945876121521,
      "eval_runtime": 12.1608,
      "eval_samples_per_second": 308.368,
      "eval_steps_per_second": 4.852,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.1314725875854492,
      "learning_rate": 0.00025029481906207064,
      "loss": 0.1832,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8704,
      "eval_loss": 0.46086084842681885,
      "eval_runtime": 10.9998,
      "eval_samples_per_second": 340.915,
      "eval_steps_per_second": 5.364,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.022194135934114456,
      "learning_rate": 0.00015460158573036287,
      "loss": 0.0836,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8874666666666666,
      "eval_loss": 0.43838638067245483,
      "eval_runtime": 10.8113,
      "eval_samples_per_second": 346.86,
      "eval_steps_per_second": 5.457,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.05780168995261192,
      "learning_rate": 7.343189618242854e-05,
      "loss": 0.0249,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9090666666666667,
      "eval_loss": 0.35258856415748596,
      "eval_runtime": 11.2945,
      "eval_samples_per_second": 332.021,
      "eval_steps_per_second": 5.224,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.015896210446953773,
      "learning_rate": 1.9143099846872597e-05,
      "loss": 0.0072,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.912,
      "eval_loss": 0.36646702885627747,
      "eval_runtime": 10.9839,
      "eval_samples_per_second": 341.409,
      "eval_steps_per_second": 5.372,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.003705956507474184,
      "learning_rate": 1.7383661915504333e-10,
      "loss": 0.0029,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9136,
      "eval_loss": 0.34945935010910034,
      "eval_runtime": 11.7476,
      "eval_samples_per_second": 319.214,
      "eval_steps_per_second": 5.022,
      "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
}