trainer_state.json

{
  "best_metric": 0.9005333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0529/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": 45.48067092895508,
      "learning_rate": 0.0004444444444444444,
      "loss": 1.2466,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8402666666666667,
      "eval_loss": 0.5247781276702881,
      "eval_runtime": 8.1634,
      "eval_samples_per_second": 459.369,
      "eval_steps_per_second": 7.227,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 23.257619857788086,
      "learning_rate": 0.0003888888888888889,
      "loss": 0.3891,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8664,
      "eval_loss": 0.45595821738243103,
      "eval_runtime": 8.1797,
      "eval_samples_per_second": 458.452,
      "eval_steps_per_second": 7.213,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 31.636688232421875,
      "learning_rate": 0.0003333333333333333,
      "loss": 0.2138,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8722666666666666,
      "eval_loss": 0.4243644177913666,
      "eval_runtime": 8.4986,
      "eval_samples_per_second": 441.251,
      "eval_steps_per_second": 6.942,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 40.29389572143555,
      "learning_rate": 0.0002777777777777778,
      "loss": 0.1323,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8784,
      "eval_loss": 0.4429454505443573,
      "eval_runtime": 7.9632,
      "eval_samples_per_second": 470.914,
      "eval_steps_per_second": 7.409,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 20.374797821044922,
      "learning_rate": 0.0002222222222222222,
      "loss": 0.0746,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8869333333333334,
      "eval_loss": 0.45559394359588623,
      "eval_runtime": 7.8594,
      "eval_samples_per_second": 477.136,
      "eval_steps_per_second": 7.507,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 44.88517761230469,
      "learning_rate": 0.00016666666666666666,
      "loss": 0.0474,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8949333333333334,
      "eval_loss": 0.43474510312080383,
      "eval_runtime": 9.7988,
      "eval_samples_per_second": 382.699,
      "eval_steps_per_second": 6.021,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 65.98124694824219,
      "learning_rate": 0.0001111111111111111,
      "loss": 0.0258,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8949333333333334,
      "eval_loss": 0.4683508574962616,
      "eval_runtime": 7.9621,
      "eval_samples_per_second": 470.978,
      "eval_steps_per_second": 7.41,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 30.30707359313965,
      "learning_rate": 5.555555555555555e-05,
      "loss": 0.0147,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9005333333333333,
      "eval_loss": 0.44701042771339417,
      "eval_runtime": 8.0046,
      "eval_samples_per_second": 468.481,
      "eval_steps_per_second": 7.371,
      "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
}