trainer_state.json

{
  "best_metric": 0.8576,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0233/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": 192.03121948242188,
      "learning_rate": 8e-05,
      "loss": 2.8659,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7122666666666667,
      "eval_loss": 1.1033464670181274,
      "eval_runtime": 12.6576,
      "eval_samples_per_second": 296.264,
      "eval_steps_per_second": 4.661,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 148.57989501953125,
      "learning_rate": 6e-05,
      "loss": 0.8005,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8192,
      "eval_loss": 0.6190588474273682,
      "eval_runtime": 12.6621,
      "eval_samples_per_second": 296.159,
      "eval_steps_per_second": 4.66,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 137.85055541992188,
      "learning_rate": 4e-05,
      "loss": 0.5269,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8512,
      "eval_loss": 0.49464452266693115,
      "eval_runtime": 12.7161,
      "eval_samples_per_second": 294.902,
      "eval_steps_per_second": 4.64,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 166.4759063720703,
      "learning_rate": 2e-05,
      "loss": 0.4227,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8573333333333333,
      "eval_loss": 0.4679499864578247,
      "eval_runtime": 12.5336,
      "eval_samples_per_second": 299.196,
      "eval_steps_per_second": 4.707,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 225.7560272216797,
      "learning_rate": 0.0,
      "loss": 0.3558,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8576,
      "eval_loss": 0.45420515537261963,
      "eval_runtime": 14.5278,
      "eval_samples_per_second": 258.125,
      "eval_steps_per_second": 4.061,
      "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
}