trainer_state.json

{
  "best_metric": 0.816,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0482/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": 119.16134643554688,
      "learning_rate": 6.331559480312315e-05,
      "loss": 3.4013,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5829333333333333,
      "eval_loss": 2.0641846656799316,
      "eval_runtime": 7.5509,
      "eval_samples_per_second": 496.628,
      "eval_steps_per_second": 7.814,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 72.02572631835938,
      "learning_rate": 4.581559480312316e-05,
      "loss": 1.2504,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7725333333333333,
      "eval_loss": 0.8058490753173828,
      "eval_runtime": 7.4684,
      "eval_samples_per_second": 502.114,
      "eval_steps_per_second": 7.9,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 147.09844970703125,
      "learning_rate": 2.4184405196876842e-05,
      "loss": 0.7467,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8002666666666667,
      "eval_loss": 0.6662527322769165,
      "eval_runtime": 7.4611,
      "eval_samples_per_second": 502.608,
      "eval_steps_per_second": 7.908,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 100.4068603515625,
      "learning_rate": 6.684405196876842e-06,
      "loss": 0.618,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8088,
      "eval_loss": 0.6197147965431213,
      "eval_runtime": 7.4813,
      "eval_samples_per_second": 501.25,
      "eval_steps_per_second": 7.886,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 151.04498291015625,
      "learning_rate": 0.0,
      "loss": 0.5801,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.816,
      "eval_loss": 0.6131516695022583,
      "eval_runtime": 7.5495,
      "eval_samples_per_second": 496.722,
      "eval_steps_per_second": 7.815,
      "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
}