trainer_state.json

{
  "best_metric": 0.04854436591267586,
  "best_model_checkpoint": "/workspace/training_output/pack-types-vit-large-patch16-224-in21k-v3/checkpoint-8480",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 8480,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.29,
      "learning_rate": 1.8820754716981133e-05,
      "loss": 0.3074,
      "step": 500
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.7641509433962265e-05,
      "loss": 0.1848,
      "step": 1000
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.6462264150943397e-05,
      "loss": 0.1662,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9736842105263158,
      "eval_loss": 0.10735838860273361,
      "eval_runtime": 163.8582,
      "eval_samples_per_second": 14.61,
      "eval_steps_per_second": 1.831,
      "step": 1696
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.5283018867924532e-05,
      "loss": 0.1433,
      "step": 2000
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.4103773584905662e-05,
      "loss": 0.1042,
      "step": 2500
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.2924528301886794e-05,
      "loss": 0.116,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9816207184628237,
      "eval_loss": 0.054496005177497864,
      "eval_runtime": 156.052,
      "eval_samples_per_second": 15.341,
      "eval_steps_per_second": 1.922,
      "step": 3392
    },
    {
      "epoch": 2.06,
      "learning_rate": 1.1745283018867926e-05,
      "loss": 0.1292,
      "step": 3500
    },
    {
      "epoch": 2.36,
      "learning_rate": 1.0566037735849058e-05,
      "loss": 0.0945,
      "step": 4000
    },
    {
      "epoch": 2.65,
      "learning_rate": 9.38679245283019e-06,
      "loss": 0.0819,
      "step": 4500
    },
    {
      "epoch": 2.95,
      "learning_rate": 8.207547169811321e-06,
      "loss": 0.091,
      "step": 5000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9849624060150376,
      "eval_loss": 0.0647253543138504,
      "eval_runtime": 156.9067,
      "eval_samples_per_second": 15.257,
      "eval_steps_per_second": 1.912,
      "step": 5088
    },
    {
      "epoch": 3.24,
      "learning_rate": 7.028301886792454e-06,
      "loss": 0.0599,
      "step": 5500
    },
    {
      "epoch": 3.54,
      "learning_rate": 5.849056603773585e-06,
      "loss": 0.0727,
      "step": 6000
    },
    {
      "epoch": 3.83,
      "learning_rate": 4.6698113207547175e-06,
      "loss": 0.0601,
      "step": 6500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9887218045112782,
      "eval_loss": 0.05016344040632248,
      "eval_runtime": 171.2498,
      "eval_samples_per_second": 13.98,
      "eval_steps_per_second": 1.752,
      "step": 6784
    },
    {
      "epoch": 4.13,
      "learning_rate": 3.4905660377358493e-06,
      "loss": 0.0571,
      "step": 7000
    },
    {
      "epoch": 4.42,
      "learning_rate": 2.3113207547169815e-06,
      "loss": 0.0582,
      "step": 7500
    },
    {
      "epoch": 4.72,
      "learning_rate": 1.1320754716981133e-06,
      "loss": 0.0438,
      "step": 8000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9908103592314118,
      "eval_loss": 0.04854436591267586,
      "eval_runtime": 158.1591,
      "eval_samples_per_second": 15.137,
      "eval_steps_per_second": 1.897,
      "step": 8480
    },
    {
      "epoch": 5.0,
      "step": 8480,
      "total_flos": 1.857741142343353e+19,
      "train_loss": 0.023235709937113637,
      "train_runtime": 2437.0648,
      "train_samples_per_second": 27.827,
      "train_steps_per_second": 3.48
    }
  ],
  "logging_steps": 500,
  "max_steps": 8480,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.857741142343353e+19,
  "trial_name": null,
  "trial_params": null
}