trainer_state.json

{
  "best_metric": 0.7397333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0766/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 93.27885437011719,
      "learning_rate": 2.7135254915624213e-05,
      "loss": 3.8035,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4653333333333333,
      "eval_loss": 3.5754995346069336,
      "eval_runtime": 8.7344,
      "eval_samples_per_second": 429.338,
      "eval_steps_per_second": 6.755,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 338.2911376953125,
      "learning_rate": 1.963525491562421e-05,
      "loss": 3.0535,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6368,
      "eval_loss": 2.567131757736206,
      "eval_runtime": 8.5416,
      "eval_samples_per_second": 439.026,
      "eval_steps_per_second": 6.907,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 157.5355224609375,
      "learning_rate": 1.036474508437579e-05,
      "loss": 1.9607,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7085333333333333,
      "eval_loss": 1.5176113843917847,
      "eval_runtime": 8.6019,
      "eval_samples_per_second": 435.952,
      "eval_steps_per_second": 6.859,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 625.4348754882812,
      "learning_rate": 2.86474508437579e-06,
      "loss": 1.3769,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7322666666666666,
      "eval_loss": 1.3532618284225464,
      "eval_runtime": 8.5102,
      "eval_samples_per_second": 440.65,
      "eval_steps_per_second": 6.933,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 119.53907012939453,
      "learning_rate": 0.0,
      "loss": 1.244,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7397333333333334,
      "eval_loss": 1.2391868829727173,
      "eval_runtime": 8.5264,
      "eval_samples_per_second": 439.811,
      "eval_steps_per_second": 6.92,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}