trainer_state.json

{
  "best_metric": 0.9034666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0183/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 26.324567794799805,
      "learning_rate": 0.00047524221697560476,
      "loss": 1.2387,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8357333333333333,
      "eval_loss": 0.552323579788208,
      "eval_runtime": 12.3256,
      "eval_samples_per_second": 304.245,
      "eval_steps_per_second": 4.787,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 36.948543548583984,
      "learning_rate": 0.0004058724504646834,
      "loss": 0.387,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8685333333333334,
      "eval_loss": 0.4251638948917389,
      "eval_runtime": 12.3964,
      "eval_samples_per_second": 302.508,
      "eval_steps_per_second": 4.759,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 33.02044677734375,
      "learning_rate": 0.0003056302334890786,
      "loss": 0.2015,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8805333333333333,
      "eval_loss": 0.4243977963924408,
      "eval_runtime": 12.5228,
      "eval_samples_per_second": 299.453,
      "eval_steps_per_second": 4.711,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 57.166595458984375,
      "learning_rate": 0.00019436976651092142,
      "loss": 0.0929,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8901333333333333,
      "eval_loss": 0.4390522241592407,
      "eval_runtime": 12.4933,
      "eval_samples_per_second": 300.162,
      "eval_steps_per_second": 4.723,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 45.65321350097656,
      "learning_rate": 9.412754953531663e-05,
      "loss": 0.0369,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8984,
      "eval_loss": 0.45595648884773254,
      "eval_runtime": 12.4405,
      "eval_samples_per_second": 301.434,
      "eval_steps_per_second": 4.743,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 38.175392150878906,
      "learning_rate": 2.4757783024395242e-05,
      "loss": 0.0214,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8997333333333334,
      "eval_loss": 0.4488094747066498,
      "eval_runtime": 12.2457,
      "eval_samples_per_second": 306.23,
      "eval_steps_per_second": 4.818,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 32.86421585083008,
      "learning_rate": 0.0,
      "loss": 0.0183,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9034666666666666,
      "eval_loss": 0.4415113925933838,
      "eval_runtime": 12.3455,
      "eval_samples_per_second": 303.753,
      "eval_steps_per_second": 4.779,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}