trainer_state.json

{
  "best_metric": 0.8928,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0086/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": 57.56599807739258,
      "learning_rate": 0.0002909538931178862,
      "loss": 1.7183,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8184,
      "eval_loss": 0.6202251315116882,
      "eval_runtime": 8.8385,
      "eval_samples_per_second": 424.281,
      "eval_steps_per_second": 6.675,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 26.08493423461914,
      "learning_rate": 0.00026490666646784665,
      "loss": 0.5307,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8464,
      "eval_loss": 0.4969119429588318,
      "eval_runtime": 8.4372,
      "eval_samples_per_second": 444.458,
      "eval_steps_per_second": 6.993,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 39.451316833496094,
      "learning_rate": 0.000225,
      "loss": 0.3338,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8688,
      "eval_loss": 0.428274929523468,
      "eval_runtime": 8.4511,
      "eval_samples_per_second": 443.731,
      "eval_steps_per_second": 6.981,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 28.39192008972168,
      "learning_rate": 0.00017604722665003956,
      "loss": 0.2288,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8784,
      "eval_loss": 0.4172998070716858,
      "eval_runtime": 8.4098,
      "eval_samples_per_second": 445.909,
      "eval_steps_per_second": 7.016,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 20.229299545288086,
      "learning_rate": 0.00012395277334996044,
      "loss": 0.1603,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8845333333333333,
      "eval_loss": 0.41561245918273926,
      "eval_runtime": 8.4414,
      "eval_samples_per_second": 444.239,
      "eval_steps_per_second": 6.989,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 33.03364562988281,
      "learning_rate": 7.500000000000002e-05,
      "loss": 0.1194,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8832,
      "eval_loss": 0.4143126606941223,
      "eval_runtime": 8.2376,
      "eval_samples_per_second": 455.228,
      "eval_steps_per_second": 7.162,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 60.52226638793945,
      "learning_rate": 3.509333353215331e-05,
      "loss": 0.0837,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8917333333333334,
      "eval_loss": 0.4026779234409332,
      "eval_runtime": 8.4231,
      "eval_samples_per_second": 445.205,
      "eval_steps_per_second": 7.005,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 20.622360229492188,
      "learning_rate": 9.046106882113751e-06,
      "loss": 0.0611,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8928,
      "eval_loss": 0.39753350615501404,
      "eval_runtime": 9.1008,
      "eval_samples_per_second": 412.054,
      "eval_steps_per_second": 6.483,
      "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
}