trainer_state.json

{
  "best_metric": 0.9021333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0586/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 78.14249420166016,
      "learning_rate": 0.0001,
      "loss": 2.6865,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7762666666666667,
      "eval_loss": 0.8541441559791565,
      "eval_runtime": 7.4168,
      "eval_samples_per_second": 505.609,
      "eval_steps_per_second": 7.955,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 245.15736389160156,
      "learning_rate": 0.0001,
      "loss": 0.6524,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8568,
      "eval_loss": 0.4808918237686157,
      "eval_runtime": 7.3987,
      "eval_samples_per_second": 506.844,
      "eval_steps_per_second": 7.974,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 142.42657470703125,
      "learning_rate": 0.0001,
      "loss": 0.4019,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8826666666666667,
      "eval_loss": 0.40055331587791443,
      "eval_runtime": 7.4233,
      "eval_samples_per_second": 505.163,
      "eval_steps_per_second": 7.948,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 50.49427795410156,
      "learning_rate": 0.0001,
      "loss": 0.2731,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8930666666666667,
      "eval_loss": 0.3725573420524597,
      "eval_runtime": 7.5143,
      "eval_samples_per_second": 499.049,
      "eval_steps_per_second": 7.852,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 145.91831970214844,
      "learning_rate": 0.0001,
      "loss": 0.1918,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9005333333333333,
      "eval_loss": 0.3526654839515686,
      "eval_runtime": 7.3767,
      "eval_samples_per_second": 508.36,
      "eval_steps_per_second": 7.998,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 50.43217849731445,
      "learning_rate": 0.0001,
      "loss": 0.1399,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9021333333333333,
      "eval_loss": 0.363288015127182,
      "eval_runtime": 7.3781,
      "eval_samples_per_second": 508.263,
      "eval_steps_per_second": 7.997,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}