trainer_state.json

{
  "best_metric": 0.8424,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0703/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": 103.52034759521484,
      "learning_rate": 5e-05,
      "loss": 3.7142,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4392,
      "eval_loss": 3.2363271713256836,
      "eval_runtime": 15.2304,
      "eval_samples_per_second": 246.218,
      "eval_steps_per_second": 3.874,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 355.4201354980469,
      "learning_rate": 5e-05,
      "loss": 1.9529,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7072,
      "eval_loss": 1.1236692667007446,
      "eval_runtime": 15.3186,
      "eval_samples_per_second": 244.8,
      "eval_steps_per_second": 3.852,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 179.67416381835938,
      "learning_rate": 5e-05,
      "loss": 0.9247,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7946666666666666,
      "eval_loss": 0.736920177936554,
      "eval_runtime": 15.1472,
      "eval_samples_per_second": 247.571,
      "eval_steps_per_second": 3.895,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 103.25131225585938,
      "learning_rate": 5e-05,
      "loss": 0.6533,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8296,
      "eval_loss": 0.5805777311325073,
      "eval_runtime": 15.2305,
      "eval_samples_per_second": 246.217,
      "eval_steps_per_second": 3.874,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 59.49850845336914,
      "learning_rate": 5e-05,
      "loss": 0.5185,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8424,
      "eval_loss": 0.5359927415847778,
      "eval_runtime": 15.121,
      "eval_samples_per_second": 247.999,
      "eval_steps_per_second": 3.902,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}