trainer_state.json

{
  "best_metric": 0.9301333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0827/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": 87.31919860839844,
      "learning_rate": 0.0004,
      "loss": 1.1558,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8594666666666667,
      "eval_loss": 0.45758405327796936,
      "eval_runtime": 7.9738,
      "eval_samples_per_second": 470.289,
      "eval_steps_per_second": 7.399,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 33.419227600097656,
      "learning_rate": 0.0003,
      "loss": 0.2966,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9053333333333333,
      "eval_loss": 0.3064389228820801,
      "eval_runtime": 7.9303,
      "eval_samples_per_second": 472.872,
      "eval_steps_per_second": 7.44,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 24.677034378051758,
      "learning_rate": 0.0002,
      "loss": 0.1257,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9176,
      "eval_loss": 0.3103218078613281,
      "eval_runtime": 7.6922,
      "eval_samples_per_second": 487.505,
      "eval_steps_per_second": 7.67,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 18.897565841674805,
      "learning_rate": 0.0001,
      "loss": 0.0555,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9290666666666667,
      "eval_loss": 0.26680052280426025,
      "eval_runtime": 7.7366,
      "eval_samples_per_second": 484.711,
      "eval_steps_per_second": 7.626,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 33.99905776977539,
      "learning_rate": 0.0,
      "loss": 0.0253,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9301333333333334,
      "eval_loss": 0.2802744209766388,
      "eval_runtime": 7.6381,
      "eval_samples_per_second": 490.961,
      "eval_steps_per_second": 7.724,
      "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
}