trainer_state.json

{
  "best_metric": 0.8701333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0025/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": 175.46902465820312,
      "learning_rate": 8.657457896300791e-05,
      "loss": 3.1017,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6474666666666666,
      "eval_loss": 1.3335820436477661,
      "eval_runtime": 7.9364,
      "eval_samples_per_second": 472.506,
      "eval_steps_per_second": 7.434,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 106.01046752929688,
      "learning_rate": 7.681980515339464e-05,
      "loss": 0.8945,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.808,
      "eval_loss": 0.6287822127342224,
      "eval_runtime": 7.808,
      "eval_samples_per_second": 480.274,
      "eval_steps_per_second": 7.556,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 123.08013916015625,
      "learning_rate": 6.222075445642905e-05,
      "loss": 0.548,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8464,
      "eval_loss": 0.5017239451408386,
      "eval_runtime": 7.6642,
      "eval_samples_per_second": 489.286,
      "eval_steps_per_second": 7.698,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 115.2101821899414,
      "learning_rate": 4.5e-05,
      "loss": 0.4148,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8592,
      "eval_loss": 0.4695819020271301,
      "eval_runtime": 7.8964,
      "eval_samples_per_second": 474.902,
      "eval_steps_per_second": 7.472,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 54.028724670410156,
      "learning_rate": 2.7779245543570963e-05,
      "loss": 0.339,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8664,
      "eval_loss": 0.4337269067764282,
      "eval_runtime": 8.1168,
      "eval_samples_per_second": 462.004,
      "eval_steps_per_second": 7.269,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 84.24478912353516,
      "learning_rate": 1.3180194846605365e-05,
      "loss": 0.2903,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8674666666666667,
      "eval_loss": 0.4336676299571991,
      "eval_runtime": 8.0244,
      "eval_samples_per_second": 467.325,
      "eval_steps_per_second": 7.353,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 186.37371826171875,
      "learning_rate": 3.425421036992097e-06,
      "loss": 0.2679,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.864,
      "eval_loss": 0.4346645772457123,
      "eval_runtime": 7.918,
      "eval_samples_per_second": 473.606,
      "eval_steps_per_second": 7.451,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 68.9564208984375,
      "learning_rate": 0.0,
      "loss": 0.2585,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8701333333333333,
      "eval_loss": 0.4270596206188202,
      "eval_runtime": 8.1039,
      "eval_samples_per_second": 462.738,
      "eval_steps_per_second": 7.28,
      "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
}