trainer_state.json

{
  "best_metric": 0.9008,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0455/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": 165.98606872558594,
      "learning_rate": 9.504844339512095e-05,
      "loss": 2.6494,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.78,
      "eval_loss": 0.8596524596214294,
      "eval_runtime": 7.2887,
      "eval_samples_per_second": 514.492,
      "eval_steps_per_second": 8.095,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 50.62409210205078,
      "learning_rate": 8.117449009293668e-05,
      "loss": 0.5859,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8637333333333334,
      "eval_loss": 0.468138724565506,
      "eval_runtime": 7.3012,
      "eval_samples_per_second": 513.612,
      "eval_steps_per_second": 8.081,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 110.70602416992188,
      "learning_rate": 6.112604669781572e-05,
      "loss": 0.3436,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8842666666666666,
      "eval_loss": 0.3940659463405609,
      "eval_runtime": 7.2576,
      "eval_samples_per_second": 516.7,
      "eval_steps_per_second": 8.129,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 121.04878997802734,
      "learning_rate": 3.887395330218429e-05,
      "loss": 0.2195,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8936,
      "eval_loss": 0.3756085932254791,
      "eval_runtime": 7.2389,
      "eval_samples_per_second": 518.037,
      "eval_steps_per_second": 8.15,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 93.4177474975586,
      "learning_rate": 1.8825509907063327e-05,
      "loss": 0.1491,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8978666666666667,
      "eval_loss": 0.3731721341609955,
      "eval_runtime": 7.2257,
      "eval_samples_per_second": 518.981,
      "eval_steps_per_second": 8.165,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 116.06294250488281,
      "learning_rate": 4.951556604879048e-06,
      "loss": 0.1151,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9008,
      "eval_loss": 0.363692969083786,
      "eval_runtime": 7.1832,
      "eval_samples_per_second": 522.053,
      "eval_steps_per_second": 8.214,
      "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
}