trainer_state.json

{
  "best_metric": 0.8802666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0530/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": 304.3556823730469,
      "learning_rate": 7.714285714285714e-05,
      "loss": 2.9684,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6989333333333333,
      "eval_loss": 1.1684645414352417,
      "eval_runtime": 5.4454,
      "eval_samples_per_second": 688.651,
      "eval_steps_per_second": 10.835,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 77.52046966552734,
      "learning_rate": 6.428571428571429e-05,
      "loss": 0.8187,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8216,
      "eval_loss": 0.5993229150772095,
      "eval_runtime": 5.5251,
      "eval_samples_per_second": 678.715,
      "eval_steps_per_second": 10.678,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 92.13040161132812,
      "learning_rate": 5.142857142857143e-05,
      "loss": 0.5334,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8472,
      "eval_loss": 0.48329269886016846,
      "eval_runtime": 5.4263,
      "eval_samples_per_second": 691.083,
      "eval_steps_per_second": 10.873,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 124.91905975341797,
      "learning_rate": 3.857142857142857e-05,
      "loss": 0.4036,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8642666666666666,
      "eval_loss": 0.4417882561683655,
      "eval_runtime": 8.0689,
      "eval_samples_per_second": 464.745,
      "eval_steps_per_second": 7.312,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 44.56376266479492,
      "learning_rate": 2.5714285714285714e-05,
      "loss": 0.3273,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8768,
      "eval_loss": 0.3942687213420868,
      "eval_runtime": 5.4344,
      "eval_samples_per_second": 690.054,
      "eval_steps_per_second": 10.857,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 91.54936981201172,
      "learning_rate": 1.2857142857142857e-05,
      "loss": 0.2957,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8802666666666666,
      "eval_loss": 0.3845222294330597,
      "eval_runtime": 5.4396,
      "eval_samples_per_second": 689.391,
      "eval_steps_per_second": 10.846,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}