trainer_state.json

{
  "best_metric": 0.8658666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0445/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": 92.72799682617188,
      "learning_rate": 8.728616793536588e-05,
      "loss": 3.0392,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6618666666666667,
      "eval_loss": 1.2950851917266846,
      "eval_runtime": 12.2878,
      "eval_samples_per_second": 305.182,
      "eval_steps_per_second": 4.802,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 70.57789611816406,
      "learning_rate": 7.947199994035401e-05,
      "loss": 0.8844,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8042666666666667,
      "eval_loss": 0.6510961055755615,
      "eval_runtime": 8.9435,
      "eval_samples_per_second": 419.301,
      "eval_steps_per_second": 6.597,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 138.8865966796875,
      "learning_rate": 6.75e-05,
      "loss": 0.5509,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8384,
      "eval_loss": 0.5316166281700134,
      "eval_runtime": 8.436,
      "eval_samples_per_second": 444.525,
      "eval_steps_per_second": 6.994,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 104.3333511352539,
      "learning_rate": 5.2814167995011876e-05,
      "loss": 0.4133,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8549333333333333,
      "eval_loss": 0.4810153841972351,
      "eval_runtime": 8.9115,
      "eval_samples_per_second": 420.807,
      "eval_steps_per_second": 6.621,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 114.62911987304688,
      "learning_rate": 3.7185832004988136e-05,
      "loss": 0.33,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8581333333333333,
      "eval_loss": 0.4425714313983917,
      "eval_runtime": 8.7692,
      "eval_samples_per_second": 427.635,
      "eval_steps_per_second": 6.728,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 44.37010955810547,
      "learning_rate": 2.250000000000001e-05,
      "loss": 0.2747,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8602666666666666,
      "eval_loss": 0.44880664348602295,
      "eval_runtime": 8.3873,
      "eval_samples_per_second": 447.105,
      "eval_steps_per_second": 7.034,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 114.603271484375,
      "learning_rate": 1.0528000059645995e-05,
      "loss": 0.2453,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8645333333333334,
      "eval_loss": 0.4327811300754547,
      "eval_runtime": 8.3124,
      "eval_samples_per_second": 451.134,
      "eval_steps_per_second": 7.098,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 169.3743133544922,
      "learning_rate": 2.7138320646341257e-06,
      "loss": 0.2212,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8658666666666667,
      "eval_loss": 0.44701650738716125,
      "eval_runtime": 8.6646,
      "eval_samples_per_second": 432.797,
      "eval_steps_per_second": 6.809,
      "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
}