trainer_state.json

{
  "best_metric": 0.9133333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0783/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 24.04900360107422,
      "learning_rate": 0.00042857142857142855,
      "loss": 1.1668,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8669333333333333,
      "eval_loss": 0.44341301918029785,
      "eval_runtime": 13.6772,
      "eval_samples_per_second": 274.179,
      "eval_steps_per_second": 4.314,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 29.725589752197266,
      "learning_rate": 0.00035714285714285714,
      "loss": 0.3447,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.88,
      "eval_loss": 0.3922005891799927,
      "eval_runtime": 13.5525,
      "eval_samples_per_second": 276.702,
      "eval_steps_per_second": 4.353,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 24.048507690429688,
      "learning_rate": 0.0002857142857142857,
      "loss": 0.1758,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8896,
      "eval_loss": 0.39341527223587036,
      "eval_runtime": 14.3562,
      "eval_samples_per_second": 261.212,
      "eval_steps_per_second": 4.11,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 23.55982208251953,
      "learning_rate": 0.00021428571428571427,
      "loss": 0.0833,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8973333333333333,
      "eval_loss": 0.38748371601104736,
      "eval_runtime": 13.3333,
      "eval_samples_per_second": 281.25,
      "eval_steps_per_second": 4.425,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 30.73633575439453,
      "learning_rate": 0.00014285714285714284,
      "loss": 0.0487,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9101333333333333,
      "eval_loss": 0.3664014935493469,
      "eval_runtime": 13.4957,
      "eval_samples_per_second": 277.866,
      "eval_steps_per_second": 4.372,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 27.952289581298828,
      "learning_rate": 7.142857142857142e-05,
      "loss": 0.0231,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9133333333333333,
      "eval_loss": 0.36764922738075256,
      "eval_runtime": 13.2856,
      "eval_samples_per_second": 282.261,
      "eval_steps_per_second": 4.441,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}