trainer_state.json

{
  "best_metric": 0.6248,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0520/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 125.49504852294922,
      "learning_rate": 2.4e-05,
      "loss": 3.8593,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2784,
      "eval_loss": 3.7576048374176025,
      "eval_runtime": 8.6928,
      "eval_samples_per_second": 431.392,
      "eval_steps_per_second": 6.787,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 183.9536895751953,
      "learning_rate": 1.8e-05,
      "loss": 3.4594,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4597333333333333,
      "eval_loss": 3.175330877304077,
      "eval_runtime": 9.0817,
      "eval_samples_per_second": 412.917,
      "eval_steps_per_second": 6.497,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 85.1482162475586,
      "learning_rate": 1.2e-05,
      "loss": 2.6822,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5504,
      "eval_loss": 2.368213653564453,
      "eval_runtime": 8.7221,
      "eval_samples_per_second": 429.944,
      "eval_steps_per_second": 6.764,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 330.4733581542969,
      "learning_rate": 6e-06,
      "loss": 2.0422,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6162666666666666,
      "eval_loss": 1.8575851917266846,
      "eval_runtime": 8.7481,
      "eval_samples_per_second": 428.662,
      "eval_steps_per_second": 6.744,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 274.441162109375,
      "learning_rate": 0.0,
      "loss": 1.7737,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6248,
      "eval_loss": 1.8262245655059814,
      "eval_runtime": 9.1894,
      "eval_samples_per_second": 408.081,
      "eval_steps_per_second": 6.42,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}