trainer_state.json

{
  "best_metric": 0.9096,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0896/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": 35.52180480957031,
      "learning_rate": 0.0003,
      "loss": 1.4166,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8714666666666666,
      "eval_loss": 0.42606765031814575,
      "eval_runtime": 12.7141,
      "eval_samples_per_second": 294.949,
      "eval_steps_per_second": 4.641,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 32.54019546508789,
      "learning_rate": 0.0003,
      "loss": 0.3314,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8912,
      "eval_loss": 0.3743750751018524,
      "eval_runtime": 12.7006,
      "eval_samples_per_second": 295.261,
      "eval_steps_per_second": 4.645,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 53.38423156738281,
      "learning_rate": 0.0003,
      "loss": 0.1857,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9026666666666666,
      "eval_loss": 0.3539322316646576,
      "eval_runtime": 12.9194,
      "eval_samples_per_second": 290.262,
      "eval_steps_per_second": 4.567,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 50.010501861572266,
      "learning_rate": 0.0003,
      "loss": 0.1148,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9032,
      "eval_loss": 0.37535130977630615,
      "eval_runtime": 12.8299,
      "eval_samples_per_second": 292.285,
      "eval_steps_per_second": 4.599,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 21.175752639770508,
      "learning_rate": 0.0003,
      "loss": 0.0844,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.904,
      "eval_loss": 0.40016353130340576,
      "eval_runtime": 12.7906,
      "eval_samples_per_second": 293.183,
      "eval_steps_per_second": 4.613,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 25.689027786254883,
      "learning_rate": 0.0003,
      "loss": 0.0629,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9096,
      "eval_loss": 0.41143015027046204,
      "eval_runtime": 13.1989,
      "eval_samples_per_second": 284.114,
      "eval_steps_per_second": 4.47,
      "step": 1998
    }
  ],
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}