trainer_state.json

{
  "best_metric": 0.952443857331572,
  "best_model_checkpoint": "vit-base-patch16-224-finetuned-landscape-test/checkpoint-115",
  "epoch": 4.842105263157895,
  "eval_steps": 500,
  "global_step": 115,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.42105263157894735,
      "grad_norm": 4.562849998474121,
      "learning_rate": 4.166666666666667e-05,
      "loss": 1.7494,
      "step": 10
    },
    {
      "epoch": 0.8421052631578947,
      "grad_norm": 1.9164351224899292,
      "learning_rate": 4.611650485436894e-05,
      "loss": 0.8077,
      "step": 20
    },
    {
      "epoch": 0.968421052631579,
      "eval_accuracy": 0.9273447820343461,
      "eval_loss": 0.2953506112098694,
      "eval_runtime": 4.958,
      "eval_samples_per_second": 152.681,
      "eval_steps_per_second": 4.841,
      "step": 23
    },
    {
      "epoch": 1.263157894736842,
      "grad_norm": 2.033348798751831,
      "learning_rate": 4.12621359223301e-05,
      "loss": 0.3319,
      "step": 30
    },
    {
      "epoch": 1.6842105263157894,
      "grad_norm": 1.7747331857681274,
      "learning_rate": 3.6407766990291265e-05,
      "loss": 0.2325,
      "step": 40
    },
    {
      "epoch": 1.9789473684210526,
      "eval_accuracy": 0.9445178335535006,
      "eval_loss": 0.16748179495334625,
      "eval_runtime": 5.0695,
      "eval_samples_per_second": 149.324,
      "eval_steps_per_second": 4.734,
      "step": 47
    },
    {
      "epoch": 2.1052631578947367,
      "grad_norm": 1.1569668054580688,
      "learning_rate": 3.155339805825243e-05,
      "loss": 0.2244,
      "step": 50
    },
    {
      "epoch": 2.526315789473684,
      "grad_norm": 1.468355655670166,
      "learning_rate": 2.6699029126213593e-05,
      "loss": 0.1922,
      "step": 60
    },
    {
      "epoch": 2.9473684210526314,
      "grad_norm": 1.393397569656372,
      "learning_rate": 2.1844660194174756e-05,
      "loss": 0.1987,
      "step": 70
    },
    {
      "epoch": 2.9894736842105263,
      "eval_accuracy": 0.9511228533685601,
      "eval_loss": 0.16012056171894073,
      "eval_runtime": 5.0061,
      "eval_samples_per_second": 151.215,
      "eval_steps_per_second": 4.794,
      "step": 71
    },
    {
      "epoch": 3.3684210526315788,
      "grad_norm": 1.0821892023086548,
      "learning_rate": 1.6990291262135926e-05,
      "loss": 0.1473,
      "step": 80
    },
    {
      "epoch": 3.7894736842105265,
      "grad_norm": 0.9723425507545471,
      "learning_rate": 1.2135922330097088e-05,
      "loss": 0.1641,
      "step": 90
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9498018494055482,
      "eval_loss": 0.15651127696037292,
      "eval_runtime": 5.088,
      "eval_samples_per_second": 148.782,
      "eval_steps_per_second": 4.717,
      "step": 95
    },
    {
      "epoch": 4.2105263157894735,
      "grad_norm": 1.3188167810440063,
      "learning_rate": 7.281553398058253e-06,
      "loss": 0.1634,
      "step": 100
    },
    {
      "epoch": 4.631578947368421,
      "grad_norm": 0.9832009673118591,
      "learning_rate": 2.4271844660194174e-06,
      "loss": 0.1497,
      "step": 110
    },
    {
      "epoch": 4.842105263157895,
      "eval_accuracy": 0.952443857331572,
      "eval_loss": 0.15358571708202362,
      "eval_runtime": 5.0096,
      "eval_samples_per_second": 151.111,
      "eval_steps_per_second": 4.791,
      "step": 115
    },
    {
      "epoch": 4.842105263157895,
      "step": 115,
      "total_flos": 1.1370032325653299e+18,
      "train_loss": 0.3839595815409785,
      "train_runtime": 264.0032,
      "train_samples_per_second": 57.348,
      "train_steps_per_second": 0.436
    }
  ],
  "logging_steps": 10,
  "max_steps": 115,
  "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": 1.1370032325653299e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}