{
  "best_metric": 0.9431968295904888,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finalterm/checkpoint-142",
  "epoch": 9.68421052631579,
  "eval_steps": 500,
  "global_step": 230,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.42105263157894735,
      "grad_norm": 3.932831048965454,
      "learning_rate": 2.173913043478261e-05,
      "loss": 1.7278,
      "step": 10
    },
    {
      "epoch": 0.8421052631578947,
      "grad_norm": 4.1621294021606445,
      "learning_rate": 4.347826086956522e-05,
      "loss": 1.263,
      "step": 20
    },
    {
      "epoch": 0.968421052631579,
      "eval_accuracy": 0.8731836195508587,
      "eval_loss": 0.5365312099456787,
      "eval_runtime": 3.2064,
      "eval_samples_per_second": 236.094,
      "eval_steps_per_second": 7.485,
      "step": 23
    },
    {
      "epoch": 1.263157894736842,
      "grad_norm": 5.375840187072754,
      "learning_rate": 4.830917874396135e-05,
      "loss": 0.5541,
      "step": 30
    },
    {
      "epoch": 1.6842105263157894,
      "grad_norm": 6.915546417236328,
      "learning_rate": 4.589371980676328e-05,
      "loss": 0.3861,
      "step": 40
    },
    {
      "epoch": 1.9789473684210526,
      "eval_accuracy": 0.9194187582562747,
      "eval_loss": 0.21320690214633942,
      "eval_runtime": 3.1377,
      "eval_samples_per_second": 241.261,
      "eval_steps_per_second": 7.649,
      "step": 47
    },
    {
      "epoch": 2.1052631578947367,
      "grad_norm": 6.825714111328125,
      "learning_rate": 4.347826086956522e-05,
      "loss": 0.3799,
      "step": 50
    },
    {
      "epoch": 2.526315789473684,
      "grad_norm": 4.591681957244873,
      "learning_rate": 4.106280193236715e-05,
      "loss": 0.2617,
      "step": 60
    },
    {
      "epoch": 2.9473684210526314,
      "grad_norm": 4.161971569061279,
      "learning_rate": 3.864734299516908e-05,
      "loss": 0.3085,
      "step": 70
    },
    {
      "epoch": 2.9894736842105263,
      "eval_accuracy": 0.9194187582562747,
      "eval_loss": 0.20058244466781616,
      "eval_runtime": 3.1669,
      "eval_samples_per_second": 239.038,
      "eval_steps_per_second": 7.578,
      "step": 71
    },
    {
      "epoch": 3.3684210526315788,
      "grad_norm": 4.695600986480713,
      "learning_rate": 3.6231884057971014e-05,
      "loss": 0.2462,
      "step": 80
    },
    {
      "epoch": 3.7894736842105265,
      "grad_norm": 4.2908549308776855,
      "learning_rate": 3.381642512077295e-05,
      "loss": 0.2545,
      "step": 90
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9339498018494056,
      "eval_loss": 0.18688541650772095,
      "eval_runtime": 3.1019,
      "eval_samples_per_second": 244.04,
      "eval_steps_per_second": 7.737,
      "step": 95
    },
    {
      "epoch": 4.2105263157894735,
      "grad_norm": 4.509713172912598,
      "learning_rate": 3.140096618357488e-05,
      "loss": 0.2281,
      "step": 100
    },
    {
      "epoch": 4.631578947368421,
      "grad_norm": 7.101438999176025,
      "learning_rate": 2.8985507246376814e-05,
      "loss": 0.2407,
      "step": 110
    },
    {
      "epoch": 4.968421052631579,
      "eval_accuracy": 0.9392338177014531,
      "eval_loss": 0.17513839900493622,
      "eval_runtime": 3.111,
      "eval_samples_per_second": 243.333,
      "eval_steps_per_second": 7.715,
      "step": 118
    },
    {
      "epoch": 5.052631578947368,
      "grad_norm": 7.366419792175293,
      "learning_rate": 2.6570048309178748e-05,
      "loss": 0.2298,
      "step": 120
    },
    {
      "epoch": 5.473684210526316,
      "grad_norm": 8.756230354309082,
      "learning_rate": 2.4154589371980676e-05,
      "loss": 0.2282,
      "step": 130
    },
    {
      "epoch": 5.894736842105263,
      "grad_norm": 3.7817883491516113,
      "learning_rate": 2.173913043478261e-05,
      "loss": 0.2092,
      "step": 140
    },
    {
      "epoch": 5.978947368421053,
      "eval_accuracy": 0.9431968295904888,
      "eval_loss": 0.16813714802265167,
      "eval_runtime": 3.1144,
      "eval_samples_per_second": 243.068,
      "eval_steps_per_second": 7.706,
      "step": 142
    },
    {
      "epoch": 6.315789473684211,
      "grad_norm": 3.032972574234009,
      "learning_rate": 1.932367149758454e-05,
      "loss": 0.1912,
      "step": 150
    },
    {
      "epoch": 6.7368421052631575,
      "grad_norm": 3.4733083248138428,
      "learning_rate": 1.6908212560386476e-05,
      "loss": 0.1941,
      "step": 160
    },
    {
      "epoch": 6.989473684210527,
      "eval_accuracy": 0.9365918097754293,
      "eval_loss": 0.16664038598537445,
      "eval_runtime": 3.1139,
      "eval_samples_per_second": 243.101,
      "eval_steps_per_second": 7.707,
      "step": 166
    },
    {
      "epoch": 7.157894736842105,
      "grad_norm": 3.442589282989502,
      "learning_rate": 1.4492753623188407e-05,
      "loss": 0.1849,
      "step": 170
    },
    {
      "epoch": 7.578947368421053,
      "grad_norm": 3.700207471847534,
      "learning_rate": 1.2077294685990338e-05,
      "loss": 0.1907,
      "step": 180
    },
    {
      "epoch": 8.0,
      "grad_norm": 4.851794719696045,
      "learning_rate": 9.66183574879227e-06,
      "loss": 0.1758,
      "step": 190
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.940554821664465,
      "eval_loss": 0.16963057219982147,
      "eval_runtime": 3.1313,
      "eval_samples_per_second": 241.75,
      "eval_steps_per_second": 7.664,
      "step": 190
    },
    {
      "epoch": 8.421052631578947,
      "grad_norm": 4.217889785766602,
      "learning_rate": 7.246376811594203e-06,
      "loss": 0.1544,
      "step": 200
    },
    {
      "epoch": 8.842105263157894,
      "grad_norm": 5.522456169128418,
      "learning_rate": 4.830917874396135e-06,
      "loss": 0.1846,
      "step": 210
    },
    {
      "epoch": 8.968421052631578,
      "eval_accuracy": 0.9326287978863936,
      "eval_loss": 0.1659202128648758,
      "eval_runtime": 3.0925,
      "eval_samples_per_second": 244.784,
      "eval_steps_per_second": 7.761,
      "step": 213
    },
    {
      "epoch": 9.263157894736842,
      "grad_norm": 2.8615903854370117,
      "learning_rate": 2.4154589371980677e-06,
      "loss": 0.1599,
      "step": 220
    },
    {
      "epoch": 9.68421052631579,
      "grad_norm": 4.822436332702637,
      "learning_rate": 0.0,
      "loss": 0.1825,
      "step": 230
    },
    {
      "epoch": 9.68421052631579,
      "eval_accuracy": 0.9365918097754293,
      "eval_loss": 0.16516228020191193,
      "eval_runtime": 3.1267,
      "eval_samples_per_second": 242.106,
      "eval_steps_per_second": 7.676,
      "step": 230
    },
    {
      "epoch": 9.68421052631579,
      "step": 230,
      "total_flos": 7.291573574754632e+17,
      "train_loss": 0.35373894960983937,
      "train_runtime": 310.9576,
      "train_samples_per_second": 97.377,
      "train_steps_per_second": 0.74
    }
  ],
  "logging_steps": 10,
  "max_steps": 230,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 7.291573574754632e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}