trainer_state.json

{
  "best_metric": 0.8866666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0817/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": 123.5261001586914,
      "learning_rate": 8e-05,
      "loss": 3.0437,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6712,
      "eval_loss": 1.3257086277008057,
      "eval_runtime": 8.6411,
      "eval_samples_per_second": 433.974,
      "eval_steps_per_second": 6.828,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 149.60385131835938,
      "learning_rate": 7.000000000000001e-05,
      "loss": 0.8861,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8152,
      "eval_loss": 0.619427502155304,
      "eval_runtime": 7.9041,
      "eval_samples_per_second": 474.438,
      "eval_steps_per_second": 7.464,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 81.6816177368164,
      "learning_rate": 6e-05,
      "loss": 0.5519,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8544,
      "eval_loss": 0.4847291111946106,
      "eval_runtime": 7.9565,
      "eval_samples_per_second": 471.313,
      "eval_steps_per_second": 7.415,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 217.58436584472656,
      "learning_rate": 5e-05,
      "loss": 0.425,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8661333333333333,
      "eval_loss": 0.4444667100906372,
      "eval_runtime": 7.9184,
      "eval_samples_per_second": 473.581,
      "eval_steps_per_second": 7.451,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 42.883174896240234,
      "learning_rate": 4e-05,
      "loss": 0.3633,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.40252408385276794,
      "eval_runtime": 7.804,
      "eval_samples_per_second": 480.525,
      "eval_steps_per_second": 7.56,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 186.01976013183594,
      "learning_rate": 3e-05,
      "loss": 0.3075,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8821333333333333,
      "eval_loss": 0.3913953900337219,
      "eval_runtime": 7.8691,
      "eval_samples_per_second": 476.547,
      "eval_steps_per_second": 7.498,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 37.740318298339844,
      "learning_rate": 2e-05,
      "loss": 0.2686,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8781333333333333,
      "eval_loss": 0.39314499497413635,
      "eval_runtime": 7.842,
      "eval_samples_per_second": 478.192,
      "eval_steps_per_second": 7.524,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 69.27288818359375,
      "learning_rate": 1e-05,
      "loss": 0.2492,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8866666666666667,
      "eval_loss": 0.3813925087451935,
      "eval_runtime": 7.865,
      "eval_samples_per_second": 476.795,
      "eval_steps_per_second": 7.502,
      "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
}