trainer_state.json

{
  "best_metric": 0.043642520904541016,
  "best_model_checkpoint": "/workspace/training_output/pack-types-vit-huge-patch14-224-in21k/checkpoint-8380",
  "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": 0.7754,
      "step": 500
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.7613365155131266e-05,
      "loss": 0.3257,
      "step": 1000
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.6420047732696897e-05,
      "loss": 0.2292,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9754860524091293,
      "eval_loss": 0.12931108474731445,
      "eval_runtime": 199.7614,
      "eval_samples_per_second": 11.844,
      "eval_steps_per_second": 1.482,
      "step": 1676
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.5226730310262532e-05,
      "loss": 0.1825,
      "step": 2000
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.4033412887828164e-05,
      "loss": 0.1507,
      "step": 2500
    },
    {
      "epoch": 1.79,
      "learning_rate": 1.2840095465393797e-05,
      "loss": 0.1376,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9826711749788672,
      "eval_loss": 0.07693611830472946,
      "eval_runtime": 199.5492,
      "eval_samples_per_second": 11.857,
      "eval_steps_per_second": 1.483,
      "step": 3352
    },
    {
      "epoch": 2.09,
      "learning_rate": 1.1646778042959427e-05,
      "loss": 0.1307,
      "step": 3500
    },
    {
      "epoch": 2.39,
      "learning_rate": 1.045346062052506e-05,
      "loss": 0.1099,
      "step": 4000
    },
    {
      "epoch": 2.68,
      "learning_rate": 9.260143198090693e-06,
      "loss": 0.1186,
      "step": 4500
    },
    {
      "epoch": 2.98,
      "learning_rate": 8.066825775656326e-06,
      "loss": 0.1122,
      "step": 5000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.985207100591716,
      "eval_loss": 0.056492872536182404,
      "eval_runtime": 201.7734,
      "eval_samples_per_second": 11.726,
      "eval_steps_per_second": 1.467,
      "step": 5028
    },
    {
      "epoch": 3.28,
      "learning_rate": 6.873508353221957e-06,
      "loss": 0.0818,
      "step": 5500
    },
    {
      "epoch": 3.58,
      "learning_rate": 5.68019093078759e-06,
      "loss": 0.0971,
      "step": 6000
    },
    {
      "epoch": 3.88,
      "learning_rate": 4.486873508353222e-06,
      "loss": 0.0759,
      "step": 6500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9873203719357565,
      "eval_loss": 0.05010140314698219,
      "eval_runtime": 199.0051,
      "eval_samples_per_second": 11.889,
      "eval_steps_per_second": 1.487,
      "step": 6704
    },
    {
      "epoch": 4.18,
      "learning_rate": 3.293556085918855e-06,
      "loss": 0.0833,
      "step": 7000
    },
    {
      "epoch": 4.47,
      "learning_rate": 2.100238663484487e-06,
      "loss": 0.073,
      "step": 7500
    },
    {
      "epoch": 4.77,
      "learning_rate": 9.069212410501194e-07,
      "loss": 0.0678,
      "step": 8000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.989010989010989,
      "eval_loss": 0.043642520904541016,
      "eval_runtime": 202.3495,
      "eval_samples_per_second": 11.693,
      "eval_steps_per_second": 1.463,
      "step": 8380
    },
    {
      "epoch": 5.0,
      "step": 8380,
      "total_flos": 3.818394112407598e+19,
      "train_loss": 0.16767553475136404,
      "train_runtime": 9223.4668,
      "train_samples_per_second": 7.267,
      "train_steps_per_second": 0.909
    }
  ],
  "logging_steps": 500,
  "max_steps": 8380,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3.818394112407598e+19,
  "trial_name": null,
  "trial_params": null
}