trainer_state.json

{
  "best_metric": 0.8632,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0564/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": 77.82820129394531,
      "learning_rate": 0.0003,
      "loss": 1.5848,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8034666666666667,
      "eval_loss": 0.6461576223373413,
      "eval_runtime": 7.5123,
      "eval_samples_per_second": 499.179,
      "eval_steps_per_second": 7.854,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 48.974822998046875,
      "learning_rate": 0.0003,
      "loss": 0.4705,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8373333333333334,
      "eval_loss": 0.5306445956230164,
      "eval_runtime": 7.4608,
      "eval_samples_per_second": 502.624,
      "eval_steps_per_second": 7.908,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 29.686079025268555,
      "learning_rate": 0.0003,
      "loss": 0.2752,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8629333333333333,
      "eval_loss": 0.48718926310539246,
      "eval_runtime": 7.444,
      "eval_samples_per_second": 503.76,
      "eval_steps_per_second": 7.926,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 28.941404342651367,
      "learning_rate": 0.0003,
      "loss": 0.1883,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8578666666666667,
      "eval_loss": 0.5159169435501099,
      "eval_runtime": 7.4203,
      "eval_samples_per_second": 505.371,
      "eval_steps_per_second": 7.951,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 36.005332946777344,
      "learning_rate": 0.0003,
      "loss": 0.121,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8632,
      "eval_loss": 0.5453027486801147,
      "eval_runtime": 7.4308,
      "eval_samples_per_second": 504.656,
      "eval_steps_per_second": 7.94,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}