trainer_state.json

{
  "best_metric": 0.88,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0087/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": 41.1666374206543,
      "learning_rate": 0.0004375,
      "loss": 1.4244,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7888,
      "eval_loss": 0.6625471115112305,
      "eval_runtime": 8.2941,
      "eval_samples_per_second": 452.13,
      "eval_steps_per_second": 7.114,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 42.36841583251953,
      "learning_rate": 0.000375,
      "loss": 0.5332,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8256,
      "eval_loss": 0.5360684990882874,
      "eval_runtime": 8.3326,
      "eval_samples_per_second": 450.038,
      "eval_steps_per_second": 7.081,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 31.027273178100586,
      "learning_rate": 0.0003125,
      "loss": 0.3327,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8477333333333333,
      "eval_loss": 0.49341776967048645,
      "eval_runtime": 8.9697,
      "eval_samples_per_second": 418.075,
      "eval_steps_per_second": 6.578,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 24.79557991027832,
      "learning_rate": 0.00025,
      "loss": 0.2266,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.864,
      "eval_loss": 0.45885077118873596,
      "eval_runtime": 8.3455,
      "eval_samples_per_second": 449.346,
      "eval_steps_per_second": 7.07,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 23.950098037719727,
      "learning_rate": 0.0001875,
      "loss": 0.1532,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8674666666666667,
      "eval_loss": 0.45624199509620667,
      "eval_runtime": 8.3899,
      "eval_samples_per_second": 446.965,
      "eval_steps_per_second": 7.032,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 52.34629821777344,
      "learning_rate": 0.000125,
      "loss": 0.1001,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8706666666666667,
      "eval_loss": 0.459536075592041,
      "eval_runtime": 8.3636,
      "eval_samples_per_second": 448.371,
      "eval_steps_per_second": 7.054,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 15.002752304077148,
      "learning_rate": 6.25e-05,
      "loss": 0.0632,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8717333333333334,
      "eval_loss": 0.48400723934173584,
      "eval_runtime": 8.5852,
      "eval_samples_per_second": 436.799,
      "eval_steps_per_second": 6.872,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 54.799015045166016,
      "learning_rate": 0.0,
      "loss": 0.0509,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.88,
      "eval_loss": 0.455447256565094,
      "eval_runtime": 8.4054,
      "eval_samples_per_second": 446.144,
      "eval_steps_per_second": 7.019,
      "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
}