trainer_state.json

{
  "best_metric": 0.9133333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0734/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 43.45814514160156,
      "learning_rate": 0.0004849231551964771,
      "loss": 1.1901,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8464,
      "eval_loss": 0.5143876075744629,
      "eval_runtime": 15.5399,
      "eval_samples_per_second": 241.314,
      "eval_steps_per_second": 3.797,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 23.139732360839844,
      "learning_rate": 0.0004415111107797445,
      "loss": 0.3641,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8768,
      "eval_loss": 0.41975387930870056,
      "eval_runtime": 15.6803,
      "eval_samples_per_second": 239.153,
      "eval_steps_per_second": 3.763,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 20.135164260864258,
      "learning_rate": 0.000375,
      "loss": 0.1939,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8864,
      "eval_loss": 0.4175671339035034,
      "eval_runtime": 15.547,
      "eval_samples_per_second": 241.204,
      "eval_steps_per_second": 3.795,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 33.29143524169922,
      "learning_rate": 0.00029341204441673266,
      "loss": 0.1096,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8944,
      "eval_loss": 0.4239102303981781,
      "eval_runtime": 15.436,
      "eval_samples_per_second": 242.938,
      "eval_steps_per_second": 3.822,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 25.880136489868164,
      "learning_rate": 0.00020658795558326743,
      "loss": 0.0552,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8989333333333334,
      "eval_loss": 0.42714953422546387,
      "eval_runtime": 15.5018,
      "eval_samples_per_second": 241.907,
      "eval_steps_per_second": 3.806,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 40.535545349121094,
      "learning_rate": 0.00012500000000000006,
      "loss": 0.0273,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9058666666666667,
      "eval_loss": 0.430266410112381,
      "eval_runtime": 15.4549,
      "eval_samples_per_second": 242.641,
      "eval_steps_per_second": 3.818,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 23.503982543945312,
      "learning_rate": 5.848888922025553e-05,
      "loss": 0.0168,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9101333333333333,
      "eval_loss": 0.41611576080322266,
      "eval_runtime": 15.4144,
      "eval_samples_per_second": 243.279,
      "eval_steps_per_second": 3.828,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 30.85848617553711,
      "learning_rate": 1.5076844803522921e-05,
      "loss": 0.0144,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9133333333333333,
      "eval_loss": 0.4127610921859741,
      "eval_runtime": 15.9295,
      "eval_samples_per_second": 235.412,
      "eval_steps_per_second": 3.704,
      "step": 2664
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}