trainer_state.json

{
  "best_metric": 0.9232,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0941/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": 75.42523956298828,
      "learning_rate": 0.0002713525491562421,
      "loss": 1.4862,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8653333333333333,
      "eval_loss": 0.45160040259361267,
      "eval_runtime": 8.3323,
      "eval_samples_per_second": 450.058,
      "eval_steps_per_second": 7.081,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 31.97371482849121,
      "learning_rate": 0.0001963525491562421,
      "loss": 0.3868,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9010666666666667,
      "eval_loss": 0.3309440016746521,
      "eval_runtime": 8.3537,
      "eval_samples_per_second": 448.902,
      "eval_steps_per_second": 7.063,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 42.49495315551758,
      "learning_rate": 0.0001036474508437579,
      "loss": 0.2117,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9202666666666667,
      "eval_loss": 0.2846162021160126,
      "eval_runtime": 8.4172,
      "eval_samples_per_second": 445.518,
      "eval_steps_per_second": 7.009,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 137.46768188476562,
      "learning_rate": 2.8647450843757897e-05,
      "loss": 0.1158,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.92,
      "eval_loss": 0.28150269389152527,
      "eval_runtime": 10.1648,
      "eval_samples_per_second": 368.92,
      "eval_steps_per_second": 5.804,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 81.8739013671875,
      "learning_rate": 0.0,
      "loss": 0.0852,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9232,
      "eval_loss": 0.2712608277797699,
      "eval_runtime": 8.6846,
      "eval_samples_per_second": 431.801,
      "eval_steps_per_second": 6.794,
      "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
}