trainer_state.json

{
  "best_metric": 0.8986666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0511/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": 29.498300552368164,
      "learning_rate": 0.0004809698831278217,
      "loss": 1.3532,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8221333333333334,
      "eval_loss": 0.5870492458343506,
      "eval_runtime": 9.2809,
      "eval_samples_per_second": 404.055,
      "eval_steps_per_second": 6.357,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 17.232101440429688,
      "learning_rate": 0.00042677669529663686,
      "loss": 0.4396,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8370666666666666,
      "eval_loss": 0.5127251148223877,
      "eval_runtime": 8.7271,
      "eval_samples_per_second": 429.695,
      "eval_steps_per_second": 6.761,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 19.244802474975586,
      "learning_rate": 0.0003456708580912725,
      "loss": 0.271,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8717333333333334,
      "eval_loss": 0.4567953050136566,
      "eval_runtime": 9.0974,
      "eval_samples_per_second": 412.207,
      "eval_steps_per_second": 6.485,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 49.4572868347168,
      "learning_rate": 0.00025,
      "loss": 0.1545,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8773333333333333,
      "eval_loss": 0.4415339231491089,
      "eval_runtime": 8.6264,
      "eval_samples_per_second": 434.713,
      "eval_steps_per_second": 6.839,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 17.725130081176758,
      "learning_rate": 0.00015432914190872756,
      "loss": 0.0885,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8888,
      "eval_loss": 0.42563092708587646,
      "eval_runtime": 8.7121,
      "eval_samples_per_second": 430.434,
      "eval_steps_per_second": 6.772,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 33.06433868408203,
      "learning_rate": 7.322330470336314e-05,
      "loss": 0.0497,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8986666666666666,
      "eval_loss": 0.41638314723968506,
      "eval_runtime": 8.7033,
      "eval_samples_per_second": 430.871,
      "eval_steps_per_second": 6.779,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}