trainer_state.json

{
  "best_metric": 0.8872,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0713/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 80.38449096679688,
      "learning_rate": 6.788924172750679e-05,
      "loss": 3.3729,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6090666666666666,
      "eval_loss": 1.8518353700637817,
      "eval_runtime": 14.4449,
      "eval_samples_per_second": 259.606,
      "eval_steps_per_second": 4.084,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 49.83513259887695,
      "learning_rate": 6.181155550916422e-05,
      "loss": 1.0416,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8250666666666666,
      "eval_loss": 0.6236870884895325,
      "eval_runtime": 14.3619,
      "eval_samples_per_second": 261.108,
      "eval_steps_per_second": 4.108,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 126.43315124511719,
      "learning_rate": 5.2499999999999995e-05,
      "loss": 0.5732,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8613333333333333,
      "eval_loss": 0.47743842005729675,
      "eval_runtime": 14.7842,
      "eval_samples_per_second": 253.649,
      "eval_steps_per_second": 3.991,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 148.67201232910156,
      "learning_rate": 4.1077686218342566e-05,
      "loss": 0.4447,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8765333333333334,
      "eval_loss": 0.42432349920272827,
      "eval_runtime": 14.2928,
      "eval_samples_per_second": 262.37,
      "eval_steps_per_second": 4.128,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 114.70174407958984,
      "learning_rate": 2.8922313781657438e-05,
      "loss": 0.3665,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8749333333333333,
      "eval_loss": 0.41159772872924805,
      "eval_runtime": 14.2987,
      "eval_samples_per_second": 262.261,
      "eval_steps_per_second": 4.126,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 116.13157653808594,
      "learning_rate": 1.7500000000000005e-05,
      "loss": 0.3169,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8848,
      "eval_loss": 0.38044556975364685,
      "eval_runtime": 14.3146,
      "eval_samples_per_second": 261.97,
      "eval_steps_per_second": 4.122,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 101.81547546386719,
      "learning_rate": 8.188444490835772e-06,
      "loss": 0.2869,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8850666666666667,
      "eval_loss": 0.3649190068244934,
      "eval_runtime": 14.3634,
      "eval_samples_per_second": 261.081,
      "eval_steps_per_second": 4.108,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 76.96652221679688,
      "learning_rate": 2.1107582724932087e-06,
      "loss": 0.2684,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8872,
      "eval_loss": 0.37409934401512146,
      "eval_runtime": 14.2711,
      "eval_samples_per_second": 262.769,
      "eval_steps_per_second": 4.134,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}