trainer_state.json

{
  "best_metric": 0.8944,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0704/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": 258.3254699707031,
      "learning_rate": 0.0001,
      "loss": 2.7097,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7370666666666666,
      "eval_loss": 1.037542700767517,
      "eval_runtime": 7.4592,
      "eval_samples_per_second": 502.738,
      "eval_steps_per_second": 7.91,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 98.01445007324219,
      "learning_rate": 0.0001,
      "loss": 0.7328,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8336,
      "eval_loss": 0.5711523294448853,
      "eval_runtime": 7.5032,
      "eval_samples_per_second": 499.786,
      "eval_steps_per_second": 7.863,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 131.39915466308594,
      "learning_rate": 0.0001,
      "loss": 0.4648,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8664,
      "eval_loss": 0.455702006816864,
      "eval_runtime": 7.9955,
      "eval_samples_per_second": 469.014,
      "eval_steps_per_second": 7.379,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 74.53013610839844,
      "learning_rate": 0.0001,
      "loss": 0.3372,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8741333333333333,
      "eval_loss": 0.4340991675853729,
      "eval_runtime": 7.4562,
      "eval_samples_per_second": 502.935,
      "eval_steps_per_second": 7.913,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 62.1510009765625,
      "learning_rate": 0.0001,
      "loss": 0.2659,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8882666666666666,
      "eval_loss": 0.3955594301223755,
      "eval_runtime": 7.4439,
      "eval_samples_per_second": 503.77,
      "eval_steps_per_second": 7.926,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 59.55078125,
      "learning_rate": 0.0001,
      "loss": 0.2138,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8944,
      "eval_loss": 0.3673877716064453,
      "eval_runtime": 7.4583,
      "eval_samples_per_second": 502.795,
      "eval_steps_per_second": 7.911,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}