trainer_state.json

{
  "best_metric": 0.8224,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0851/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": 288.1031188964844,
      "learning_rate": 3e-05,
      "loss": 3.8576,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3152,
      "eval_loss": 3.744849681854248,
      "eval_runtime": 13.0487,
      "eval_samples_per_second": 287.385,
      "eval_steps_per_second": 4.522,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 180.81240844726562,
      "learning_rate": 3e-05,
      "loss": 3.2799,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5362666666666667,
      "eval_loss": 2.571992874145508,
      "eval_runtime": 13.0272,
      "eval_samples_per_second": 287.859,
      "eval_steps_per_second": 4.529,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 194.20521545410156,
      "learning_rate": 3e-05,
      "loss": 1.7768,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7184,
      "eval_loss": 1.1881500482559204,
      "eval_runtime": 13.2207,
      "eval_samples_per_second": 283.647,
      "eval_steps_per_second": 4.463,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 254.28936767578125,
      "learning_rate": 3e-05,
      "loss": 0.9955,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7864,
      "eval_loss": 0.7914592623710632,
      "eval_runtime": 12.9627,
      "eval_samples_per_second": 289.291,
      "eval_steps_per_second": 4.552,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 94.32122039794922,
      "learning_rate": 3e-05,
      "loss": 0.7234,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8224,
      "eval_loss": 0.6376187205314636,
      "eval_runtime": 13.0398,
      "eval_samples_per_second": 287.581,
      "eval_steps_per_second": 4.525,
      "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
}