trainer_state.json

{
  "best_metric": 0.8778666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0577/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": 114.13864135742188,
      "learning_rate": 8.140576474687264e-05,
      "loss": 2.7785,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7194666666666667,
      "eval_loss": 1.0412194728851318,
      "eval_runtime": 7.2235,
      "eval_samples_per_second": 519.14,
      "eval_steps_per_second": 8.168,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 229.95364379882812,
      "learning_rate": 5.890576474687264e-05,
      "loss": 0.7291,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8461333333333333,
      "eval_loss": 0.5188705325126648,
      "eval_runtime": 7.248,
      "eval_samples_per_second": 517.385,
      "eval_steps_per_second": 8.14,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 98.27505493164062,
      "learning_rate": 3.1094235253127374e-05,
      "loss": 0.4478,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8648,
      "eval_loss": 0.4471728503704071,
      "eval_runtime": 7.1799,
      "eval_samples_per_second": 522.293,
      "eval_steps_per_second": 8.217,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 62.922210693359375,
      "learning_rate": 8.59423525312737e-06,
      "loss": 0.3542,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8776,
      "eval_loss": 0.42902839183807373,
      "eval_runtime": 7.1875,
      "eval_samples_per_second": 521.739,
      "eval_steps_per_second": 8.209,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 79.81520080566406,
      "learning_rate": 0.0,
      "loss": 0.3211,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.41917160153388977,
      "eval_runtime": 7.5073,
      "eval_samples_per_second": 499.517,
      "eval_steps_per_second": 7.859,
      "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
}