trainer_state.json

{
  "best_metric": 0.03428684547543526,
  "best_model_checkpoint": "/workspace/training_output/pack-types-vit-large-patch16-224-in21k/checkpoint-3352",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 8380,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3,
      "learning_rate": 1.8806682577565632e-05,
      "loss": 0.2863,
      "step": 500
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.7613365155131266e-05,
      "loss": 0.1669,
      "step": 1000
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.6420047732696897e-05,
      "loss": 0.1556,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9860524091293322,
      "eval_loss": 0.04897575080394745,
      "eval_runtime": 170.5458,
      "eval_samples_per_second": 13.873,
      "eval_steps_per_second": 1.736,
      "step": 1676
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.5226730310262532e-05,
      "loss": 0.1183,
      "step": 2000
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.4033412887828164e-05,
      "loss": 0.1141,
      "step": 2500
    },
    {
      "epoch": 1.79,
      "learning_rate": 1.2840095465393797e-05,
      "loss": 0.1185,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9894336432797971,
      "eval_loss": 0.03428684547543526,
      "eval_runtime": 161.6386,
      "eval_samples_per_second": 14.638,
      "eval_steps_per_second": 1.831,
      "step": 3352
    },
    {
      "epoch": 2.09,
      "learning_rate": 1.1646778042959427e-05,
      "loss": 0.1127,
      "step": 3500
    },
    {
      "epoch": 2.39,
      "learning_rate": 1.045346062052506e-05,
      "loss": 0.0862,
      "step": 4000
    },
    {
      "epoch": 2.68,
      "learning_rate": 9.260143198090693e-06,
      "loss": 0.1032,
      "step": 4500
    },
    {
      "epoch": 2.98,
      "learning_rate": 8.066825775656326e-06,
      "loss": 0.0815,
      "step": 5000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9881656804733728,
      "eval_loss": 0.053725942969322205,
      "eval_runtime": 159.6941,
      "eval_samples_per_second": 14.816,
      "eval_steps_per_second": 1.854,
      "step": 5028
    },
    {
      "epoch": 3.28,
      "learning_rate": 6.873508353221957e-06,
      "loss": 0.0744,
      "step": 5500
    },
    {
      "epoch": 3.58,
      "learning_rate": 5.68019093078759e-06,
      "loss": 0.0793,
      "step": 6000
    },
    {
      "epoch": 3.88,
      "learning_rate": 4.486873508353222e-06,
      "loss": 0.0503,
      "step": 6500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9915469146238377,
      "eval_loss": 0.03739229962229729,
      "eval_runtime": 160.0301,
      "eval_samples_per_second": 14.785,
      "eval_steps_per_second": 1.85,
      "step": 6704
    },
    {
      "epoch": 4.18,
      "learning_rate": 3.293556085918855e-06,
      "loss": 0.0727,
      "step": 7000
    },
    {
      "epoch": 4.47,
      "learning_rate": 2.100238663484487e-06,
      "loss": 0.0466,
      "step": 7500
    },
    {
      "epoch": 4.77,
      "learning_rate": 9.069212410501194e-07,
      "loss": 0.0447,
      "step": 8000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9915469146238377,
      "eval_loss": 0.03615482896566391,
      "eval_runtime": 162.6104,
      "eval_samples_per_second": 14.55,
      "eval_steps_per_second": 1.82,
      "step": 8380
    },
    {
      "epoch": 5.0,
      "step": 8380,
      "total_flos": 1.8360872742865766e+19,
      "train_loss": 0.10461939445258894,
      "train_runtime": 6074.546,
      "train_samples_per_second": 11.034,
      "train_steps_per_second": 1.38
    }
  ],
  "logging_steps": 500,
  "max_steps": 8380,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.8360872742865766e+19,
  "trial_name": null,
  "trial_params": null
}