trainer_state.json

{
  "best_metric": 0.8808,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0198/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": 210.370849609375,
      "learning_rate": 0.0001,
      "loss": 2.6769,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7517333333333334,
      "eval_loss": 0.8971264362335205,
      "eval_runtime": 9.0689,
      "eval_samples_per_second": 413.501,
      "eval_steps_per_second": 6.506,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 184.90109252929688,
      "learning_rate": 0.0001,
      "loss": 0.667,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8434666666666667,
      "eval_loss": 0.503759503364563,
      "eval_runtime": 7.6071,
      "eval_samples_per_second": 492.963,
      "eval_steps_per_second": 7.756,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 115.67908477783203,
      "learning_rate": 0.0001,
      "loss": 0.3821,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8581333333333333,
      "eval_loss": 0.4413103759288788,
      "eval_runtime": 7.6312,
      "eval_samples_per_second": 491.401,
      "eval_steps_per_second": 7.731,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 47.9951057434082,
      "learning_rate": 0.0001,
      "loss": 0.2395,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8712,
      "eval_loss": 0.40222734212875366,
      "eval_runtime": 7.6517,
      "eval_samples_per_second": 490.089,
      "eval_steps_per_second": 7.711,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 140.40594482421875,
      "learning_rate": 0.0001,
      "loss": 0.1552,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8712,
      "eval_loss": 0.4267241358757019,
      "eval_runtime": 7.6377,
      "eval_samples_per_second": 490.988,
      "eval_steps_per_second": 7.725,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 52.56611251831055,
      "learning_rate": 0.0001,
      "loss": 0.0904,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8808,
      "eval_loss": 0.42819133400917053,
      "eval_runtime": 7.6282,
      "eval_samples_per_second": 491.595,
      "eval_steps_per_second": 7.734,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}