trainer_state.json

{
  "best_metric": 0.8728,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0693/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 81.42725372314453,
      "learning_rate": 3e-05,
      "loss": 3.8357,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.41573333333333334,
      "eval_loss": 3.6555840969085693,
      "eval_runtime": 7.755,
      "eval_samples_per_second": 483.561,
      "eval_steps_per_second": 7.608,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 90.27020263671875,
      "learning_rate": 3e-05,
      "loss": 2.9686,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6274666666666666,
      "eval_loss": 2.096496820449829,
      "eval_runtime": 7.8146,
      "eval_samples_per_second": 479.871,
      "eval_steps_per_second": 7.55,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 134.85963439941406,
      "learning_rate": 3e-05,
      "loss": 1.3794,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7725333333333333,
      "eval_loss": 0.9281657338142395,
      "eval_runtime": 7.7328,
      "eval_samples_per_second": 484.948,
      "eval_steps_per_second": 7.63,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 240.87644958496094,
      "learning_rate": 3e-05,
      "loss": 0.76,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8224,
      "eval_loss": 0.6539841294288635,
      "eval_runtime": 7.5716,
      "eval_samples_per_second": 495.273,
      "eval_steps_per_second": 7.792,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 193.7390594482422,
      "learning_rate": 3e-05,
      "loss": 0.5497,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8405333333333334,
      "eval_loss": 0.5600317716598511,
      "eval_runtime": 7.5888,
      "eval_samples_per_second": 494.149,
      "eval_steps_per_second": 7.775,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 426.5158996582031,
      "learning_rate": 3e-05,
      "loss": 0.431,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8586666666666667,
      "eval_loss": 0.4741799831390381,
      "eval_runtime": 7.9462,
      "eval_samples_per_second": 471.922,
      "eval_steps_per_second": 7.425,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 172.79246520996094,
      "learning_rate": 3e-05,
      "loss": 0.3459,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8677333333333334,
      "eval_loss": 0.44412630796432495,
      "eval_runtime": 7.5977,
      "eval_samples_per_second": 493.571,
      "eval_steps_per_second": 7.766,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 203.3938446044922,
      "learning_rate": 3e-05,
      "loss": 0.2882,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8728,
      "eval_loss": 0.43409839272499084,
      "eval_runtime": 7.5893,
      "eval_samples_per_second": 494.114,
      "eval_steps_per_second": 7.774,
      "step": 2664
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}