trainer_state.json

{
  "best_metric": 0.8336,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0886/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": 190.030517578125,
      "learning_rate": 2.625e-05,
      "loss": 3.8395,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.44053333333333333,
      "eval_loss": 3.7300004959106445,
      "eval_runtime": 8.9795,
      "eval_samples_per_second": 417.618,
      "eval_steps_per_second": 6.571,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 116.64234161376953,
      "learning_rate": 2.25e-05,
      "loss": 3.2491,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5965333333333334,
      "eval_loss": 2.721337080001831,
      "eval_runtime": 7.8837,
      "eval_samples_per_second": 475.665,
      "eval_steps_per_second": 7.484,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 127.51585388183594,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 1.9822,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7357333333333334,
      "eval_loss": 1.4236493110656738,
      "eval_runtime": 8.0086,
      "eval_samples_per_second": 468.247,
      "eval_steps_per_second": 7.367,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 91.05712890625,
      "learning_rate": 1.5e-05,
      "loss": 1.1931,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7837333333333333,
      "eval_loss": 0.951380729675293,
      "eval_runtime": 7.9113,
      "eval_samples_per_second": 474.004,
      "eval_steps_per_second": 7.458,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 112.31963348388672,
      "learning_rate": 1.125e-05,
      "loss": 0.8838,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8130666666666667,
      "eval_loss": 0.7594538927078247,
      "eval_runtime": 8.0873,
      "eval_samples_per_second": 463.69,
      "eval_steps_per_second": 7.295,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 235.3500213623047,
      "learning_rate": 7.5e-06,
      "loss": 0.7458,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8256,
      "eval_loss": 0.6799284219741821,
      "eval_runtime": 8.0244,
      "eval_samples_per_second": 467.324,
      "eval_steps_per_second": 7.353,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 258.9883117675781,
      "learning_rate": 3.75e-06,
      "loss": 0.6585,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.832,
      "eval_loss": 0.6537559032440186,
      "eval_runtime": 8.1518,
      "eval_samples_per_second": 460.023,
      "eval_steps_per_second": 7.238,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 169.8813018798828,
      "learning_rate": 0.0,
      "loss": 0.6227,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8336,
      "eval_loss": 0.6378756761550903,
      "eval_runtime": 8.1002,
      "eval_samples_per_second": 462.954,
      "eval_steps_per_second": 7.284,
      "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
}