trainer_state.json

{
  "best_metric": 0.912,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0614/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": 104.98017120361328,
      "learning_rate": 0.000288581929876693,
      "loss": 1.4728,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8306666666666667,
      "eval_loss": 0.5624468922615051,
      "eval_runtime": 13.8241,
      "eval_samples_per_second": 271.266,
      "eval_steps_per_second": 4.268,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 56.8989372253418,
      "learning_rate": 0.00025606601717798207,
      "loss": 0.4007,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8802666666666666,
      "eval_loss": 0.40907731652259827,
      "eval_runtime": 13.7691,
      "eval_samples_per_second": 272.348,
      "eval_steps_per_second": 4.285,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 62.37528991699219,
      "learning_rate": 0.00020740251485476345,
      "loss": 0.2323,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8952,
      "eval_loss": 0.364244282245636,
      "eval_runtime": 13.8648,
      "eval_samples_per_second": 270.469,
      "eval_steps_per_second": 4.255,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 32.2694206237793,
      "learning_rate": 0.00015,
      "loss": 0.1516,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8989333333333334,
      "eval_loss": 0.35946908593177795,
      "eval_runtime": 14.1069,
      "eval_samples_per_second": 265.828,
      "eval_steps_per_second": 4.182,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 47.819313049316406,
      "learning_rate": 9.259748514523653e-05,
      "loss": 0.0804,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9013333333333333,
      "eval_loss": 0.3641752004623413,
      "eval_runtime": 14.7016,
      "eval_samples_per_second": 255.074,
      "eval_steps_per_second": 4.013,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 44.18015670776367,
      "learning_rate": 4.3933982822017876e-05,
      "loss": 0.0509,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9088,
      "eval_loss": 0.35221099853515625,
      "eval_runtime": 14.6468,
      "eval_samples_per_second": 256.029,
      "eval_steps_per_second": 4.028,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 35.07679748535156,
      "learning_rate": 1.1418070123306989e-05,
      "loss": 0.0359,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9090666666666667,
      "eval_loss": 0.3414585292339325,
      "eval_runtime": 13.815,
      "eval_samples_per_second": 271.444,
      "eval_steps_per_second": 4.271,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 35.45474624633789,
      "learning_rate": 0.0,
      "loss": 0.0307,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.912,
      "eval_loss": 0.3562934100627899,
      "eval_runtime": 13.8774,
      "eval_samples_per_second": 270.224,
      "eval_steps_per_second": 4.252,
      "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
}