trainer_state.json

{
  "best_metric": 0.8709333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0737/checkpoints/checkpoint-2997",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 2997,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 36.037532806396484,
      "learning_rate": 0.0005,
      "loss": 1.3918,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8298666666666666,
      "eval_loss": 0.5770149827003479,
      "eval_runtime": 13.2928,
      "eval_samples_per_second": 282.107,
      "eval_steps_per_second": 4.438,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 31.293607711791992,
      "learning_rate": 0.0005,
      "loss": 0.5149,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8490666666666666,
      "eval_loss": 0.5023054480552673,
      "eval_runtime": 13.2783,
      "eval_samples_per_second": 282.416,
      "eval_steps_per_second": 4.443,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 23.787630081176758,
      "learning_rate": 0.0005,
      "loss": 0.361,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8488,
      "eval_loss": 0.5284984111785889,
      "eval_runtime": 13.9408,
      "eval_samples_per_second": 268.994,
      "eval_steps_per_second": 4.232,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 25.495235443115234,
      "learning_rate": 0.0005,
      "loss": 0.2683,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.86,
      "eval_loss": 0.4904082119464874,
      "eval_runtime": 13.9259,
      "eval_samples_per_second": 269.282,
      "eval_steps_per_second": 4.237,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 22.8808650970459,
      "learning_rate": 0.0005,
      "loss": 0.2079,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8624,
      "eval_loss": 0.48976296186447144,
      "eval_runtime": 13.5718,
      "eval_samples_per_second": 276.308,
      "eval_steps_per_second": 4.347,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 21.762502670288086,
      "learning_rate": 0.0005,
      "loss": 0.1608,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8616,
      "eval_loss": 0.5315013527870178,
      "eval_runtime": 13.8552,
      "eval_samples_per_second": 270.657,
      "eval_steps_per_second": 4.258,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 30.535409927368164,
      "learning_rate": 0.0005,
      "loss": 0.1383,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8648,
      "eval_loss": 0.557260274887085,
      "eval_runtime": 13.5521,
      "eval_samples_per_second": 276.71,
      "eval_steps_per_second": 4.354,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 11.978509902954102,
      "learning_rate": 0.0005,
      "loss": 0.1126,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8629333333333333,
      "eval_loss": 0.577491819858551,
      "eval_runtime": 13.1249,
      "eval_samples_per_second": 285.716,
      "eval_steps_per_second": 4.495,
      "step": 2664
    },
    {
      "epoch": 9.0,
      "grad_norm": 20.98592185974121,
      "learning_rate": 0.0005,
      "loss": 0.101,
      "step": 2997
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8709333333333333,
      "eval_loss": 0.5323702692985535,
      "eval_runtime": 13.2141,
      "eval_samples_per_second": 283.788,
      "eval_steps_per_second": 4.465,
      "step": 2997
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.3587863155968e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}