trainer_state.json

{
  "best_metric": 0.8962666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0966/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 84.70071411132812,
      "learning_rate": 0.0001,
      "loss": 2.8265,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7362666666666666,
      "eval_loss": 1.020922064781189,
      "eval_runtime": 7.6643,
      "eval_samples_per_second": 489.281,
      "eval_steps_per_second": 7.698,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 100.85954284667969,
      "learning_rate": 0.0001,
      "loss": 0.7277,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8552,
      "eval_loss": 0.5138413310050964,
      "eval_runtime": 7.7158,
      "eval_samples_per_second": 486.018,
      "eval_steps_per_second": 7.647,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 32.55278778076172,
      "learning_rate": 0.0001,
      "loss": 0.4277,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8762666666666666,
      "eval_loss": 0.4271785616874695,
      "eval_runtime": 7.716,
      "eval_samples_per_second": 486.006,
      "eval_steps_per_second": 7.646,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 42.00246810913086,
      "learning_rate": 0.0001,
      "loss": 0.3012,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8869333333333334,
      "eval_loss": 0.3927765488624573,
      "eval_runtime": 7.7499,
      "eval_samples_per_second": 483.879,
      "eval_steps_per_second": 7.613,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 123.86675262451172,
      "learning_rate": 0.0001,
      "loss": 0.2205,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8962666666666667,
      "eval_loss": 0.36575767397880554,
      "eval_runtime": 7.5985,
      "eval_samples_per_second": 493.518,
      "eval_steps_per_second": 7.765,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}