trainer_state.json

{
  "best_metric": 0.9018666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0669/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": 43.815345764160156,
      "learning_rate": 0.0004522542485937369,
      "loss": 1.388,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8237333333333333,
      "eval_loss": 0.6145352721214294,
      "eval_runtime": 9.1192,
      "eval_samples_per_second": 411.22,
      "eval_steps_per_second": 6.47,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 25.627147674560547,
      "learning_rate": 0.00032725424859373687,
      "loss": 0.4511,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8624,
      "eval_loss": 0.47806575894355774,
      "eval_runtime": 8.8452,
      "eval_samples_per_second": 423.96,
      "eval_steps_per_second": 6.67,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 37.463584899902344,
      "learning_rate": 0.00017274575140626317,
      "loss": 0.2516,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8882666666666666,
      "eval_loss": 0.40136829018592834,
      "eval_runtime": 8.6124,
      "eval_samples_per_second": 435.417,
      "eval_steps_per_second": 6.851,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 32.15886688232422,
      "learning_rate": 4.7745751406263163e-05,
      "loss": 0.1392,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8997333333333334,
      "eval_loss": 0.3688678443431854,
      "eval_runtime": 8.6276,
      "eval_samples_per_second": 434.65,
      "eval_steps_per_second": 6.838,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 27.222192764282227,
      "learning_rate": 0.0,
      "loss": 0.0908,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9018666666666667,
      "eval_loss": 0.352429062128067,
      "eval_runtime": 8.772,
      "eval_samples_per_second": 427.495,
      "eval_steps_per_second": 6.726,
      "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
}