trainer_state.json

{
  "best_metric": 0.9104,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0250/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": 30.364980697631836,
      "learning_rate": 0.0004809698831278217,
      "loss": 1.2351,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8485333333333334,
      "eval_loss": 0.4855446517467499,
      "eval_runtime": 13.2841,
      "eval_samples_per_second": 282.293,
      "eval_steps_per_second": 4.441,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 52.50047302246094,
      "learning_rate": 0.00042677669529663686,
      "loss": 0.3786,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8802666666666666,
      "eval_loss": 0.4052076041698456,
      "eval_runtime": 13.2511,
      "eval_samples_per_second": 282.995,
      "eval_steps_per_second": 4.452,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 45.01051712036133,
      "learning_rate": 0.0003456708580912725,
      "loss": 0.1964,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8824,
      "eval_loss": 0.4382583200931549,
      "eval_runtime": 13.2499,
      "eval_samples_per_second": 283.021,
      "eval_steps_per_second": 4.453,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 29.802291870117188,
      "learning_rate": 0.00025,
      "loss": 0.1007,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8997333333333334,
      "eval_loss": 0.3889455795288086,
      "eval_runtime": 13.2041,
      "eval_samples_per_second": 284.003,
      "eval_steps_per_second": 4.468,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 28.234989166259766,
      "learning_rate": 0.00015432914190872756,
      "loss": 0.0444,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9016,
      "eval_loss": 0.38665664196014404,
      "eval_runtime": 13.1818,
      "eval_samples_per_second": 284.482,
      "eval_steps_per_second": 4.476,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 26.991832733154297,
      "learning_rate": 7.322330470336314e-05,
      "loss": 0.0227,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9104,
      "eval_loss": 0.3932892680168152,
      "eval_runtime": 13.187,
      "eval_samples_per_second": 284.37,
      "eval_steps_per_second": 4.474,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}