trainer_state.json

{
  "best_metric": 0.8133333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0961/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": 68.0191650390625,
      "learning_rate": 2.625e-05,
      "loss": 3.8396,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3984,
      "eval_loss": 3.6925041675567627,
      "eval_runtime": 13.7362,
      "eval_samples_per_second": 273.001,
      "eval_steps_per_second": 4.295,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 167.2474822998047,
      "learning_rate": 2.25e-05,
      "loss": 3.2638,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5853333333333334,
      "eval_loss": 2.807554244995117,
      "eval_runtime": 13.7762,
      "eval_samples_per_second": 272.209,
      "eval_steps_per_second": 4.283,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 361.98388671875,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 2.008,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7010666666666666,
      "eval_loss": 1.3992661237716675,
      "eval_runtime": 13.7282,
      "eval_samples_per_second": 273.161,
      "eval_steps_per_second": 4.298,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 140.1162109375,
      "learning_rate": 1.5e-05,
      "loss": 1.1964,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7656,
      "eval_loss": 0.9664367437362671,
      "eval_runtime": 13.571,
      "eval_samples_per_second": 276.325,
      "eval_steps_per_second": 4.348,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 281.00341796875,
      "learning_rate": 1.125e-05,
      "loss": 0.9079,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7829333333333334,
      "eval_loss": 0.837743878364563,
      "eval_runtime": 13.6597,
      "eval_samples_per_second": 274.531,
      "eval_steps_per_second": 4.319,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 165.92877197265625,
      "learning_rate": 7.5e-06,
      "loss": 0.7768,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8093333333333333,
      "eval_loss": 0.7424760460853577,
      "eval_runtime": 18.416,
      "eval_samples_per_second": 203.627,
      "eval_steps_per_second": 3.204,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 613.2069702148438,
      "learning_rate": 3.75e-06,
      "loss": 0.6963,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8064,
      "eval_loss": 0.7186475396156311,
      "eval_runtime": 13.5676,
      "eval_samples_per_second": 276.394,
      "eval_steps_per_second": 4.349,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 157.03570556640625,
      "learning_rate": 0.0,
      "loss": 0.6645,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8133333333333334,
      "eval_loss": 0.7023743391036987,
      "eval_runtime": 13.3691,
      "eval_samples_per_second": 280.497,
      "eval_steps_per_second": 4.413,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}