trainer_state.json

{
  "best_metric": 0.9472,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0486/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.17218206822872162,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.6224,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9093333333333333,
      "eval_loss": 0.3074714243412018,
      "eval_runtime": 7.7902,
      "eval_samples_per_second": 481.373,
      "eval_steps_per_second": 7.574,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.039715323597192764,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.1427,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9258666666666666,
      "eval_loss": 0.2700952887535095,
      "eval_runtime": 6.7528,
      "eval_samples_per_second": 555.328,
      "eval_steps_per_second": 8.737,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.5452052354812622,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0727,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9290666666666667,
      "eval_loss": 0.2619270086288452,
      "eval_runtime": 6.772,
      "eval_samples_per_second": 553.749,
      "eval_steps_per_second": 8.712,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.014812109991908073,
      "learning_rate": 5e-05,
      "loss": 0.0364,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9376,
      "eval_loss": 0.2447633147239685,
      "eval_runtime": 7.6512,
      "eval_samples_per_second": 490.119,
      "eval_steps_per_second": 7.711,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.023919595405459404,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0173,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9456,
      "eval_loss": 0.22619491815567017,
      "eval_runtime": 7.2772,
      "eval_samples_per_second": 515.305,
      "eval_steps_per_second": 8.107,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.014747492037713528,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0063,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9442666666666667,
      "eval_loss": 0.23491181433200836,
      "eval_runtime": 7.118,
      "eval_samples_per_second": 526.835,
      "eval_steps_per_second": 8.289,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": NaN,
      "learning_rate": 3.82861996937452e-06,
      "loss": 0.0034,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9469333333333333,
      "eval_loss": 0.2292962223291397,
      "eval_runtime": 6.7773,
      "eval_samples_per_second": 553.318,
      "eval_steps_per_second": 8.706,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.011877468787133694,
      "learning_rate": 3.476732383100867e-11,
      "loss": 0.0021,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9472,
      "eval_loss": 0.22798173129558563,
      "eval_runtime": 6.9336,
      "eval_samples_per_second": 540.848,
      "eval_steps_per_second": 8.509,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}