trainer_state.json

{
  "best_metric": 0.8634666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0299/checkpoints/checkpoint-2997",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 2997,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 220.6021728515625,
      "learning_rate": 5e-05,
      "loss": 3.7099,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4514666666666667,
      "eval_loss": 3.1971657276153564,
      "eval_runtime": 8.3512,
      "eval_samples_per_second": 449.035,
      "eval_steps_per_second": 7.065,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 158.89276123046875,
      "learning_rate": 5e-05,
      "loss": 1.8726,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7128,
      "eval_loss": 1.115404725074768,
      "eval_runtime": 7.5396,
      "eval_samples_per_second": 497.373,
      "eval_steps_per_second": 7.825,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 147.91065979003906,
      "learning_rate": 5e-05,
      "loss": 0.9263,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7810666666666667,
      "eval_loss": 0.7654125094413757,
      "eval_runtime": 7.5654,
      "eval_samples_per_second": 495.681,
      "eval_steps_per_second": 7.799,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 131.35098266601562,
      "learning_rate": 5e-05,
      "loss": 0.6679,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8181333333333334,
      "eval_loss": 0.6011420488357544,
      "eval_runtime": 7.5592,
      "eval_samples_per_second": 496.084,
      "eval_steps_per_second": 7.805,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 168.19735717773438,
      "learning_rate": 5e-05,
      "loss": 0.5239,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.832,
      "eval_loss": 0.5324193239212036,
      "eval_runtime": 7.5739,
      "eval_samples_per_second": 495.124,
      "eval_steps_per_second": 7.79,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 127.90918731689453,
      "learning_rate": 5e-05,
      "loss": 0.4336,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.852,
      "eval_loss": 0.49097833037376404,
      "eval_runtime": 8.1824,
      "eval_samples_per_second": 458.298,
      "eval_steps_per_second": 7.211,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 55.3325080871582,
      "learning_rate": 5e-05,
      "loss": 0.3666,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8592,
      "eval_loss": 0.4547177851200104,
      "eval_runtime": 7.5408,
      "eval_samples_per_second": 497.291,
      "eval_steps_per_second": 7.824,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 353.92144775390625,
      "learning_rate": 5e-05,
      "loss": 0.3063,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8597333333333333,
      "eval_loss": 0.4644729793071747,
      "eval_runtime": 7.5149,
      "eval_samples_per_second": 499.012,
      "eval_steps_per_second": 7.851,
      "step": 2664
    },
    {
      "epoch": 9.0,
      "grad_norm": 182.07293701171875,
      "learning_rate": 5e-05,
      "loss": 0.2577,
      "step": 2997
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8634666666666667,
      "eval_loss": 0.43845710158348083,
      "eval_runtime": 7.4755,
      "eval_samples_per_second": 501.641,
      "eval_steps_per_second": 7.892,
      "step": 2997
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 7.3587863155968e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}