trainer_state.json

{
  "best_metric": 0.06565134227275848,
  "best_model_checkpoint": "./deit-base-mri/checkpoint-1000",
  "epoch": 2.0,
  "global_step": 1250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "learning_rate": 4.804e-05,
      "loss": 0.0316,
      "step": 50
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.604e-05,
      "loss": 0.0196,
      "step": 100
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.4040000000000005e-05,
      "loss": 0.0426,
      "step": 150
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.2040000000000004e-05,
      "loss": 0.0195,
      "step": 200
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.004e-05,
      "loss": 0.0365,
      "step": 250
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.804e-05,
      "loss": 0.0025,
      "step": 300
    },
    {
      "epoch": 0.56,
      "learning_rate": 3.604e-05,
      "loss": 0.008,
      "step": 350
    },
    {
      "epoch": 0.64,
      "learning_rate": 3.404e-05,
      "loss": 0.0139,
      "step": 400
    },
    {
      "epoch": 0.72,
      "learning_rate": 3.2040000000000005e-05,
      "loss": 0.006,
      "step": 450
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.004e-05,
      "loss": 0.0107,
      "step": 500
    },
    {
      "epoch": 0.8,
      "eval_accuracy": 0.9886524822695035,
      "eval_loss": 0.07820000499486923,
      "eval_runtime": 41.7405,
      "eval_samples_per_second": 84.45,
      "eval_steps_per_second": 10.565,
      "step": 500
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.804e-05,
      "loss": 0.0178,
      "step": 550
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.6040000000000005e-05,
      "loss": 0.0096,
      "step": 600
    },
    {
      "epoch": 1.04,
      "learning_rate": 2.404e-05,
      "loss": 0.0089,
      "step": 650
    },
    {
      "epoch": 1.12,
      "learning_rate": 2.2040000000000002e-05,
      "loss": 0.0007,
      "step": 700
    },
    {
      "epoch": 1.2,
      "learning_rate": 2.004e-05,
      "loss": 0.0001,
      "step": 750
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.804e-05,
      "loss": 0.0003,
      "step": 800
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.604e-05,
      "loss": 0.0056,
      "step": 850
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.4040000000000001e-05,
      "loss": 0.0046,
      "step": 900
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.204e-05,
      "loss": 0.0001,
      "step": 950
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.004e-05,
      "loss": 0.0065,
      "step": 1000
    },
    {
      "epoch": 1.6,
      "eval_accuracy": 0.9900709219858156,
      "eval_loss": 0.06565134227275848,
      "eval_runtime": 42.3123,
      "eval_samples_per_second": 83.309,
      "eval_steps_per_second": 10.422,
      "step": 1000
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.040000000000001e-06,
      "loss": 0.0042,
      "step": 1050
    },
    {
      "epoch": 1.76,
      "learning_rate": 6.040000000000001e-06,
      "loss": 0.0135,
      "step": 1100
    },
    {
      "epoch": 1.84,
      "learning_rate": 4.04e-06,
      "loss": 0.0052,
      "step": 1150
    },
    {
      "epoch": 1.92,
      "learning_rate": 2.0400000000000004e-06,
      "loss": 0.0005,
      "step": 1200
    },
    {
      "epoch": 2.0,
      "learning_rate": 4e-08,
      "loss": 0.0007,
      "step": 1250
    },
    {
      "epoch": 2.0,
      "step": 1250,
      "total_flos": 3.0955504328740454e+18,
      "train_loss": 0.010772173270583153,
      "train_runtime": 1053.5222,
      "train_samples_per_second": 37.917,
      "train_steps_per_second": 1.186
    }
  ],
  "max_steps": 1250,
  "num_train_epochs": 2,
  "total_flos": 3.0955504328740454e+18,
  "trial_name": null,
  "trial_params": null
}