trainer_state.json

{
  "best_metric": 0.8864,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0368/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": 45.798274993896484,
      "learning_rate": 0.0002571428571428571,
      "loss": 1.7175,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8021333333333334,
      "eval_loss": 0.6348996162414551,
      "eval_runtime": 7.8505,
      "eval_samples_per_second": 477.679,
      "eval_steps_per_second": 7.515,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 31.32755470275879,
      "learning_rate": 0.00021428571428571427,
      "loss": 0.5154,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8549333333333333,
      "eval_loss": 0.46765780448913574,
      "eval_runtime": 7.8648,
      "eval_samples_per_second": 476.808,
      "eval_steps_per_second": 7.502,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 39.13698196411133,
      "learning_rate": 0.0001714285714285714,
      "loss": 0.3228,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8610666666666666,
      "eval_loss": 0.4331819415092468,
      "eval_runtime": 7.8373,
      "eval_samples_per_second": 478.483,
      "eval_steps_per_second": 7.528,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 38.57865524291992,
      "learning_rate": 0.00012857142857142855,
      "loss": 0.2311,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8698666666666667,
      "eval_loss": 0.4319781959056854,
      "eval_runtime": 7.7574,
      "eval_samples_per_second": 483.407,
      "eval_steps_per_second": 7.606,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 54.50212860107422,
      "learning_rate": 8.57142857142857e-05,
      "loss": 0.157,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8784,
      "eval_loss": 0.42096880078315735,
      "eval_runtime": 7.7701,
      "eval_samples_per_second": 482.619,
      "eval_steps_per_second": 7.593,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 31.427003860473633,
      "learning_rate": 4.285714285714285e-05,
      "loss": 0.1156,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8864,
      "eval_loss": 0.39330121874809265,
      "eval_runtime": 7.8299,
      "eval_samples_per_second": 478.936,
      "eval_steps_per_second": 7.535,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}