trainer_state.json

{
  "best_metric": 0.8954666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0111/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 84.75235748291016,
      "learning_rate": 0.0001,
      "loss": 2.5844,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7856,
      "eval_loss": 0.8004981875419617,
      "eval_runtime": 14.2396,
      "eval_samples_per_second": 263.35,
      "eval_steps_per_second": 4.143,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 107.3158950805664,
      "learning_rate": 0.0001,
      "loss": 0.6202,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8554666666666667,
      "eval_loss": 0.48352205753326416,
      "eval_runtime": 14.6185,
      "eval_samples_per_second": 256.524,
      "eval_steps_per_second": 4.036,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 77.50785827636719,
      "learning_rate": 0.0001,
      "loss": 0.3469,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8821333333333333,
      "eval_loss": 0.3965870440006256,
      "eval_runtime": 14.2192,
      "eval_samples_per_second": 263.729,
      "eval_steps_per_second": 4.149,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 56.68440628051758,
      "learning_rate": 0.0001,
      "loss": 0.2085,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8904,
      "eval_loss": 0.37322869896888733,
      "eval_runtime": 14.118,
      "eval_samples_per_second": 265.618,
      "eval_steps_per_second": 4.179,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 50.45477294921875,
      "learning_rate": 0.0001,
      "loss": 0.1257,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8914666666666666,
      "eval_loss": 0.4024505317211151,
      "eval_runtime": 14.2052,
      "eval_samples_per_second": 263.987,
      "eval_steps_per_second": 4.153,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 79.58975982666016,
      "learning_rate": 0.0001,
      "loss": 0.0908,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8954666666666666,
      "eval_loss": 0.41121235489845276,
      "eval_runtime": 14.0816,
      "eval_samples_per_second": 266.304,
      "eval_steps_per_second": 4.19,
      "step": 1998
    }
  ],
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}