trainer_state.json

{
  "best_metric": 0.8877333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0075/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": 44.679481506347656,
      "learning_rate": 6.222222222222222e-05,
      "loss": 3.2466,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6728,
      "eval_loss": 1.4403784275054932,
      "eval_runtime": 7.7029,
      "eval_samples_per_second": 486.832,
      "eval_steps_per_second": 7.659,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 108.91317749023438,
      "learning_rate": 5.444444444444444e-05,
      "loss": 0.908,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8224,
      "eval_loss": 0.6347119808197021,
      "eval_runtime": 7.6385,
      "eval_samples_per_second": 490.937,
      "eval_steps_per_second": 7.724,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 58.48265838623047,
      "learning_rate": 4.666666666666666e-05,
      "loss": 0.5523,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8490666666666666,
      "eval_loss": 0.48779940605163574,
      "eval_runtime": 7.1993,
      "eval_samples_per_second": 520.881,
      "eval_steps_per_second": 8.195,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 82.97557067871094,
      "learning_rate": 3.8888888888888884e-05,
      "loss": 0.4165,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8661333333333333,
      "eval_loss": 0.4467521607875824,
      "eval_runtime": 7.2031,
      "eval_samples_per_second": 520.61,
      "eval_steps_per_second": 8.191,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 86.51017761230469,
      "learning_rate": 3.111111111111111e-05,
      "loss": 0.3338,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8733333333333333,
      "eval_loss": 0.41808101534843445,
      "eval_runtime": 7.1781,
      "eval_samples_per_second": 522.425,
      "eval_steps_per_second": 8.219,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 209.5661163330078,
      "learning_rate": 2.333333333333333e-05,
      "loss": 0.2916,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8744,
      "eval_loss": 0.4035494327545166,
      "eval_runtime": 7.2484,
      "eval_samples_per_second": 517.357,
      "eval_steps_per_second": 8.14,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 127.57278442382812,
      "learning_rate": 1.5555555555555555e-05,
      "loss": 0.2472,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8842666666666666,
      "eval_loss": 0.3825065791606903,
      "eval_runtime": 7.1334,
      "eval_samples_per_second": 525.698,
      "eval_steps_per_second": 8.271,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 72.5779800415039,
      "learning_rate": 7.777777777777777e-06,
      "loss": 0.2266,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8877333333333334,
      "eval_loss": 0.37383267283439636,
      "eval_runtime": 7.2032,
      "eval_samples_per_second": 520.603,
      "eval_steps_per_second": 8.191,
      "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
}