trainer_state.json

{
  "best_metric": 0.8746666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0847/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": 92.10770416259766,
      "learning_rate": 6.125e-05,
      "loss": 3.3491,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6101333333333333,
      "eval_loss": 2.003835439682007,
      "eval_runtime": 8.0504,
      "eval_samples_per_second": 465.814,
      "eval_steps_per_second": 7.329,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 154.2924041748047,
      "learning_rate": 5.2499999999999995e-05,
      "loss": 1.2308,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7941333333333334,
      "eval_loss": 0.781033992767334,
      "eval_runtime": 8.1186,
      "eval_samples_per_second": 461.904,
      "eval_steps_per_second": 7.267,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 129.52381896972656,
      "learning_rate": 4.374999999999999e-05,
      "loss": 0.7137,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8290666666666666,
      "eval_loss": 0.6225418448448181,
      "eval_runtime": 8.1455,
      "eval_samples_per_second": 460.378,
      "eval_steps_per_second": 7.243,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 197.4302520751953,
      "learning_rate": 3.5e-05,
      "loss": 0.5665,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8506666666666667,
      "eval_loss": 0.5280658602714539,
      "eval_runtime": 8.0914,
      "eval_samples_per_second": 463.452,
      "eval_steps_per_second": 7.292,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 289.75543212890625,
      "learning_rate": 2.6249999999999998e-05,
      "loss": 0.4681,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8576,
      "eval_loss": 0.49320194125175476,
      "eval_runtime": 8.0827,
      "eval_samples_per_second": 463.953,
      "eval_steps_per_second": 7.3,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 118.31321716308594,
      "learning_rate": 1.75e-05,
      "loss": 0.4263,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8717333333333334,
      "eval_loss": 0.44251134991645813,
      "eval_runtime": 8.4223,
      "eval_samples_per_second": 445.248,
      "eval_steps_per_second": 7.005,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 153.5042266845703,
      "learning_rate": 8.75e-06,
      "loss": 0.3815,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8736,
      "eval_loss": 0.4447650611400604,
      "eval_runtime": 7.9977,
      "eval_samples_per_second": 468.886,
      "eval_steps_per_second": 7.377,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 105.56419372558594,
      "learning_rate": 0.0,
      "loss": 0.3645,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8746666666666667,
      "eval_loss": 0.43222713470458984,
      "eval_runtime": 8.7628,
      "eval_samples_per_second": 427.945,
      "eval_steps_per_second": 6.733,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}