trainer_state.json

{
  "best_metric": 0.8704,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0820/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": 6.073007583618164,
      "learning_rate": 2.7135254915624213e-05,
      "loss": 2.1046,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7608,
      "eval_loss": 1.0227065086364746,
      "eval_runtime": 13.4032,
      "eval_samples_per_second": 279.785,
      "eval_steps_per_second": 4.402,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 1.9662164055386567e-05,
      "loss": 0.7553,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8213333333333334,
      "eval_loss": 0.6622294783592224,
      "eval_runtime": 13.4534,
      "eval_samples_per_second": 278.74,
      "eval_steps_per_second": 4.386,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.3489391803741455,
      "learning_rate": 1.0391670726449015e-05,
      "loss": 0.4764,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8581333333333333,
      "eval_loss": 0.5146906971931458,
      "eval_runtime": 13.4284,
      "eval_samples_per_second": 279.258,
      "eval_steps_per_second": 4.394,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.045289516448975,
      "learning_rate": 2.881402548656824e-06,
      "loss": 0.3386,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8658666666666667,
      "eval_loss": 0.45805585384368896,
      "eval_runtime": 13.2344,
      "eval_samples_per_second": 283.352,
      "eval_steps_per_second": 4.458,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 17.743131637573242,
      "learning_rate": 2.670129987614178e-11,
      "loss": 0.2796,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8704,
      "eval_loss": 0.45658183097839355,
      "eval_runtime": 14.1042,
      "eval_samples_per_second": 265.878,
      "eval_steps_per_second": 4.183,
      "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
}