trainer_state.json

{
  "best_metric": 0.9008,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0922/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": 39.22636795043945,
      "learning_rate": 0.0002713525491562421,
      "loss": 1.6473,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8274666666666667,
      "eval_loss": 0.5796734094619751,
      "eval_runtime": 9.5371,
      "eval_samples_per_second": 393.201,
      "eval_steps_per_second": 6.186,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 45.73237991333008,
      "learning_rate": 0.0001963525491562421,
      "loss": 0.4699,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.4293041527271271,
      "eval_runtime": 9.6849,
      "eval_samples_per_second": 387.202,
      "eval_steps_per_second": 6.092,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 44.27059555053711,
      "learning_rate": 0.0001036474508437579,
      "loss": 0.2743,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8872,
      "eval_loss": 0.3847120702266693,
      "eval_runtime": 9.537,
      "eval_samples_per_second": 393.207,
      "eval_steps_per_second": 6.186,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 47.562808990478516,
      "learning_rate": 2.8647450843757897e-05,
      "loss": 0.1805,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.896,
      "eval_loss": 0.35835179686546326,
      "eval_runtime": 9.596,
      "eval_samples_per_second": 390.788,
      "eval_steps_per_second": 6.148,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 62.33792495727539,
      "learning_rate": 0.0,
      "loss": 0.1385,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9008,
      "eval_loss": 0.34254714846611023,
      "eval_runtime": 9.4825,
      "eval_samples_per_second": 395.465,
      "eval_steps_per_second": 6.222,
      "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
}