trainer_state.json

{
  "best_metric": 0.9072,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0181/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": 26.471656799316406,
      "learning_rate": 0.0004849231551964771,
      "loss": 1.4101,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8325333333333333,
      "eval_loss": 0.5913702845573425,
      "eval_runtime": 5.3938,
      "eval_samples_per_second": 695.247,
      "eval_steps_per_second": 10.939,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 25.600467681884766,
      "learning_rate": 0.0004415111107797445,
      "loss": 0.4927,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8738666666666667,
      "eval_loss": 0.45427563786506653,
      "eval_runtime": 5.3793,
      "eval_samples_per_second": 697.116,
      "eval_steps_per_second": 10.968,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 26.994258880615234,
      "learning_rate": 0.000375,
      "loss": 0.3139,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8722666666666666,
      "eval_loss": 0.45097869634628296,
      "eval_runtime": 5.4122,
      "eval_samples_per_second": 692.88,
      "eval_steps_per_second": 10.901,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 36.968326568603516,
      "learning_rate": 0.00029341204441673266,
      "loss": 0.1994,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8888,
      "eval_loss": 0.42106080055236816,
      "eval_runtime": 5.3849,
      "eval_samples_per_second": 696.388,
      "eval_steps_per_second": 10.957,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 22.726430892944336,
      "learning_rate": 0.00020658795558326743,
      "loss": 0.1278,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8949333333333334,
      "eval_loss": 0.4018915891647339,
      "eval_runtime": 5.3523,
      "eval_samples_per_second": 700.627,
      "eval_steps_per_second": 11.023,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 31.23035430908203,
      "learning_rate": 0.00012500000000000006,
      "loss": 0.0818,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.904,
      "eval_loss": 0.3775927424430847,
      "eval_runtime": 5.3517,
      "eval_samples_per_second": 700.715,
      "eval_steps_per_second": 11.025,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 34.19134521484375,
      "learning_rate": 5.848888922025553e-05,
      "loss": 0.0458,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9005333333333333,
      "eval_loss": 0.385460764169693,
      "eval_runtime": 5.3706,
      "eval_samples_per_second": 698.243,
      "eval_steps_per_second": 10.986,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 28.690004348754883,
      "learning_rate": 1.5076844803522921e-05,
      "loss": 0.0348,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9072,
      "eval_loss": 0.3715668320655823,
      "eval_runtime": 5.3801,
      "eval_samples_per_second": 697.017,
      "eval_steps_per_second": 10.966,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}