trainer_state.json

{
  "best_metric": 0.9554666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0638/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": 2.77311372756958,
      "learning_rate": 2.7135254915624213e-05,
      "loss": 1.1105,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9338666666666666,
      "eval_loss": 0.2657354474067688,
      "eval_runtime": 11.7007,
      "eval_samples_per_second": 320.493,
      "eval_steps_per_second": 5.042,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3972071409225464,
      "learning_rate": 1.963525491562421e-05,
      "loss": 0.1342,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9485333333333333,
      "eval_loss": 0.18163664638996124,
      "eval_runtime": 10.7661,
      "eval_samples_per_second": 348.317,
      "eval_steps_per_second": 5.48,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.1711016744375229,
      "learning_rate": 1.036474508437579e-05,
      "loss": 0.0446,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.952,
      "eval_loss": 0.1719745695590973,
      "eval_runtime": 10.7813,
      "eval_samples_per_second": 347.826,
      "eval_steps_per_second": 5.472,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.19226647913455963,
      "learning_rate": 2.86474508437579e-06,
      "loss": 0.0183,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9549333333333333,
      "eval_loss": 0.1715882122516632,
      "eval_runtime": 10.5877,
      "eval_samples_per_second": 354.186,
      "eval_steps_per_second": 5.573,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.4463564455509186,
      "learning_rate": 0.0,
      "loss": 0.0112,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9554666666666667,
      "eval_loss": 0.17153838276863098,
      "eval_runtime": 10.7347,
      "eval_samples_per_second": 349.335,
      "eval_steps_per_second": 5.496,
      "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}