trainer_state.json

{
  "best_metric": 0.9016,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0330/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": 73.88716888427734,
      "learning_rate": 5e-05,
      "loss": 3.654,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5048,
      "eval_loss": 2.943110942840576,
      "eval_runtime": 5.4395,
      "eval_samples_per_second": 689.407,
      "eval_steps_per_second": 10.847,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 142.72274780273438,
      "learning_rate": 5e-05,
      "loss": 1.6467,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7896,
      "eval_loss": 0.8627796173095703,
      "eval_runtime": 5.4482,
      "eval_samples_per_second": 688.295,
      "eval_steps_per_second": 10.829,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 73.6986083984375,
      "learning_rate": 5e-05,
      "loss": 0.6931,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.836,
      "eval_loss": 0.5869982838630676,
      "eval_runtime": 5.4308,
      "eval_samples_per_second": 690.501,
      "eval_steps_per_second": 10.864,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 219.8760986328125,
      "learning_rate": 5e-05,
      "loss": 0.4898,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8648,
      "eval_loss": 0.4742652475833893,
      "eval_runtime": 5.4541,
      "eval_samples_per_second": 687.553,
      "eval_steps_per_second": 10.818,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 103.17538452148438,
      "learning_rate": 5e-05,
      "loss": 0.3797,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8712,
      "eval_loss": 0.42330020666122437,
      "eval_runtime": 5.4179,
      "eval_samples_per_second": 692.154,
      "eval_steps_per_second": 10.89,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 108.58040618896484,
      "learning_rate": 5e-05,
      "loss": 0.3148,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.884,
      "eval_loss": 0.3990298807621002,
      "eval_runtime": 5.4201,
      "eval_samples_per_second": 691.87,
      "eval_steps_per_second": 10.885,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 214.72467041015625,
      "learning_rate": 5e-05,
      "loss": 0.2746,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8928,
      "eval_loss": 0.3642450273036957,
      "eval_runtime": 5.4459,
      "eval_samples_per_second": 688.59,
      "eval_steps_per_second": 10.834,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 62.55629348754883,
      "learning_rate": 5e-05,
      "loss": 0.2309,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9016,
      "eval_loss": 0.344721257686615,
      "eval_runtime": 5.4403,
      "eval_samples_per_second": 689.297,
      "eval_steps_per_second": 10.845,
      "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
}