trainer_state.json

{
  "best_metric": 0.8341333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0403/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": 73.86934661865234,
      "learning_rate": 2.9095389311788626e-05,
      "loss": 3.8164,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4216,
      "eval_loss": 3.611102342605591,
      "eval_runtime": 12.7962,
      "eval_samples_per_second": 293.056,
      "eval_steps_per_second": 4.611,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 149.992431640625,
      "learning_rate": 2.649066664678467e-05,
      "loss": 2.9254,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6565333333333333,
      "eval_loss": 2.084704637527466,
      "eval_runtime": 12.6753,
      "eval_samples_per_second": 295.851,
      "eval_steps_per_second": 4.655,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 130.8927001953125,
      "learning_rate": 2.25e-05,
      "loss": 1.4381,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7581333333333333,
      "eval_loss": 1.004410982131958,
      "eval_runtime": 12.6214,
      "eval_samples_per_second": 297.113,
      "eval_steps_per_second": 4.675,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 322.0773620605469,
      "learning_rate": 1.760472266500396e-05,
      "loss": 0.8881,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7842666666666667,
      "eval_loss": 0.7588279247283936,
      "eval_runtime": 13.2907,
      "eval_samples_per_second": 282.153,
      "eval_steps_per_second": 4.439,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 234.82847595214844,
      "learning_rate": 1.2395277334996045e-05,
      "loss": 0.7101,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.824,
      "eval_loss": 0.6279023885726929,
      "eval_runtime": 12.6515,
      "eval_samples_per_second": 296.407,
      "eval_steps_per_second": 4.663,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 136.2180633544922,
      "learning_rate": 7.500000000000004e-06,
      "loss": 0.6263,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8282666666666667,
      "eval_loss": 0.6072946190834045,
      "eval_runtime": 12.6658,
      "eval_samples_per_second": 296.072,
      "eval_steps_per_second": 4.658,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 189.1455841064453,
      "learning_rate": 3.5093333532153316e-06,
      "loss": 0.5767,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8328,
      "eval_loss": 0.5663248896598816,
      "eval_runtime": 12.7236,
      "eval_samples_per_second": 294.728,
      "eval_steps_per_second": 4.637,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 136.14561462402344,
      "learning_rate": 9.046106882113753e-07,
      "loss": 0.5527,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8341333333333333,
      "eval_loss": 0.5645958185195923,
      "eval_runtime": 12.6765,
      "eval_samples_per_second": 295.823,
      "eval_steps_per_second": 4.654,
      "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
}