trainer_state.json

{
  "best_global_step": 74,
  "best_metric": 0.9348659003831418,
  "best_model_checkpoint": "efficientnet-b0-finetuned-chest-xray-pneumonia\\checkpoint-74",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 111,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.272108843537415,
      "grad_norm": 1.6204841136932373,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.6286,
      "step": 10
    },
    {
      "epoch": 0.54421768707483,
      "grad_norm": 2.213111400604248,
      "learning_rate": 4.6464646464646464e-05,
      "loss": 0.5382,
      "step": 20
    },
    {
      "epoch": 0.8163265306122449,
      "grad_norm": 1.6756197214126587,
      "learning_rate": 4.141414141414142e-05,
      "loss": 0.4634,
      "step": 30
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8505747126436781,
      "eval_loss": 0.41398224234580994,
      "eval_runtime": 32.2148,
      "eval_samples_per_second": 16.204,
      "eval_steps_per_second": 0.528,
      "step": 37
    },
    {
      "epoch": 1.0816326530612246,
      "grad_norm": 1.109337329864502,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 0.3788,
      "step": 40
    },
    {
      "epoch": 1.3537414965986394,
      "grad_norm": 1.0655840635299683,
      "learning_rate": 3.131313131313132e-05,
      "loss": 0.3463,
      "step": 50
    },
    {
      "epoch": 1.6258503401360545,
      "grad_norm": 0.9670161008834839,
      "learning_rate": 2.6262626262626268e-05,
      "loss": 0.3097,
      "step": 60
    },
    {
      "epoch": 1.8979591836734695,
      "grad_norm": 1.4370367527008057,
      "learning_rate": 2.1212121212121215e-05,
      "loss": 0.2814,
      "step": 70
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9348659003831418,
      "eval_loss": 0.22404897212982178,
      "eval_runtime": 35.1145,
      "eval_samples_per_second": 14.866,
      "eval_steps_per_second": 0.484,
      "step": 74
    },
    {
      "epoch": 2.163265306122449,
      "grad_norm": 3.019960880279541,
      "learning_rate": 1.6161616161616165e-05,
      "loss": 0.2773,
      "step": 80
    },
    {
      "epoch": 2.435374149659864,
      "grad_norm": 1.1376591920852661,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.2692,
      "step": 90
    },
    {
      "epoch": 2.707482993197279,
      "grad_norm": 1.2763034105300903,
      "learning_rate": 6.060606060606061e-06,
      "loss": 0.2478,
      "step": 100
    },
    {
      "epoch": 2.979591836734694,
      "grad_norm": 0.9628028869628906,
      "learning_rate": 1.0101010101010103e-06,
      "loss": 0.2495,
      "step": 110
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9157088122605364,
      "eval_loss": 0.24497930705547333,
      "eval_runtime": 29.7585,
      "eval_samples_per_second": 17.541,
      "eval_steps_per_second": 0.571,
      "step": 111
    },
    {
      "epoch": 3.0,
      "step": 111,
      "total_flos": 5.100223024705536e+16,
      "train_loss": 0.36133632412901867,
      "train_runtime": 2186.928,
      "train_samples_per_second": 6.439,
      "train_steps_per_second": 0.051
    }
  ],
  "logging_steps": 10,
  "max_steps": 111,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 5.100223024705536e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}