trainer_state.json

{
  "best_metric": 0.872,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0178/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": 41.35458755493164,
      "learning_rate": 0.0003,
      "loss": 1.6386,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7890666666666667,
      "eval_loss": 0.688805103302002,
      "eval_runtime": 14.8635,
      "eval_samples_per_second": 252.296,
      "eval_steps_per_second": 3.969,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 24.68541145324707,
      "learning_rate": 0.0003,
      "loss": 0.5309,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8429333333333333,
      "eval_loss": 0.4978673458099365,
      "eval_runtime": 14.6361,
      "eval_samples_per_second": 256.216,
      "eval_steps_per_second": 4.031,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 30.165016174316406,
      "learning_rate": 0.0003,
      "loss": 0.3609,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.848,
      "eval_loss": 0.49466440081596375,
      "eval_runtime": 14.7099,
      "eval_samples_per_second": 254.931,
      "eval_steps_per_second": 4.011,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 27.351362228393555,
      "learning_rate": 0.0003,
      "loss": 0.2643,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8626666666666667,
      "eval_loss": 0.47356459498405457,
      "eval_runtime": 14.732,
      "eval_samples_per_second": 254.548,
      "eval_steps_per_second": 4.005,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 38.73955154418945,
      "learning_rate": 0.0003,
      "loss": 0.1994,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8501333333333333,
      "eval_loss": 0.5338327884674072,
      "eval_runtime": 14.6413,
      "eval_samples_per_second": 256.125,
      "eval_steps_per_second": 4.03,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 21.361379623413086,
      "learning_rate": 0.0003,
      "loss": 0.1618,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8674666666666667,
      "eval_loss": 0.4919414520263672,
      "eval_runtime": 14.5997,
      "eval_samples_per_second": 256.855,
      "eval_steps_per_second": 4.041,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 37.84736633300781,
      "learning_rate": 0.0003,
      "loss": 0.1432,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8672,
      "eval_loss": 0.4971838593482971,
      "eval_runtime": 14.7228,
      "eval_samples_per_second": 254.706,
      "eval_steps_per_second": 4.007,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 27.98440170288086,
      "learning_rate": 0.0003,
      "loss": 0.1217,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.872,
      "eval_loss": 0.5117279887199402,
      "eval_runtime": 14.7487,
      "eval_samples_per_second": 254.259,
      "eval_steps_per_second": 4.0,
      "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
}