trainer_state.json

{
  "best_metric": 0.852,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0088/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": 178.75245666503906,
      "learning_rate": 3e-05,
      "loss": 3.8211,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.45866666666666667,
      "eval_loss": 3.603787422180176,
      "eval_runtime": 7.2477,
      "eval_samples_per_second": 517.406,
      "eval_steps_per_second": 8.141,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 164.03968811035156,
      "learning_rate": 3e-05,
      "loss": 2.7897,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6512,
      "eval_loss": 1.7138655185699463,
      "eval_runtime": 7.4342,
      "eval_samples_per_second": 504.428,
      "eval_steps_per_second": 7.936,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 116.86677551269531,
      "learning_rate": 3e-05,
      "loss": 1.1696,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7824,
      "eval_loss": 0.8301701545715332,
      "eval_runtime": 7.4863,
      "eval_samples_per_second": 500.913,
      "eval_steps_per_second": 7.881,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 253.57350158691406,
      "learning_rate": 3e-05,
      "loss": 0.6952,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8202666666666667,
      "eval_loss": 0.6131578087806702,
      "eval_runtime": 7.4939,
      "eval_samples_per_second": 500.408,
      "eval_steps_per_second": 7.873,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 146.0651092529297,
      "learning_rate": 3e-05,
      "loss": 0.5065,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8408,
      "eval_loss": 0.5070021152496338,
      "eval_runtime": 7.2938,
      "eval_samples_per_second": 514.133,
      "eval_steps_per_second": 8.089,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 182.35916137695312,
      "learning_rate": 3e-05,
      "loss": 0.4016,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.852,
      "eval_loss": 0.46920743584632874,
      "eval_runtime": 8.0032,
      "eval_samples_per_second": 468.562,
      "eval_steps_per_second": 7.372,
      "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
}