trainer_state.json

{
  "best_metric": 0.8501333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0950/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": 104.80438995361328,
      "learning_rate": 6.531088913245535e-05,
      "loss": 3.2579,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6565333333333333,
      "eval_loss": 1.604994297027588,
      "eval_runtime": 12.3109,
      "eval_samples_per_second": 304.609,
      "eval_steps_per_second": 4.793,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 188.67440795898438,
      "learning_rate": 5.2499999999999995e-05,
      "loss": 0.9536,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8058666666666666,
      "eval_loss": 0.6517338156700134,
      "eval_runtime": 12.7224,
      "eval_samples_per_second": 294.755,
      "eval_steps_per_second": 4.637,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 79.8405532836914,
      "learning_rate": 3.5e-05,
      "loss": 0.5283,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8341333333333333,
      "eval_loss": 0.5269869565963745,
      "eval_runtime": 12.4776,
      "eval_samples_per_second": 300.538,
      "eval_steps_per_second": 4.728,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 165.27015686035156,
      "learning_rate": 1.7500000000000005e-05,
      "loss": 0.3891,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8464,
      "eval_loss": 0.5003998875617981,
      "eval_runtime": 12.3891,
      "eval_samples_per_second": 302.685,
      "eval_steps_per_second": 4.762,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 297.12774658203125,
      "learning_rate": 4.689110867544645e-06,
      "loss": 0.3339,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8442666666666667,
      "eval_loss": 0.4852418303489685,
      "eval_runtime": 12.3981,
      "eval_samples_per_second": 302.466,
      "eval_steps_per_second": 4.759,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 110.47047424316406,
      "learning_rate": 0.0,
      "loss": 0.309,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8501333333333333,
      "eval_loss": 0.4824371039867401,
      "eval_runtime": 12.3144,
      "eval_samples_per_second": 304.522,
      "eval_steps_per_second": 4.791,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}