trainer_state.json

{
  "best_metric": 0.892,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0852/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": 113.76178741455078,
      "learning_rate": 6.125e-05,
      "loss": 3.2002,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6730666666666667,
      "eval_loss": 1.4884686470031738,
      "eval_runtime": 8.5682,
      "eval_samples_per_second": 437.665,
      "eval_steps_per_second": 6.886,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 170.7351837158203,
      "learning_rate": 5.2499999999999995e-05,
      "loss": 0.9585,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8213333333333334,
      "eval_loss": 0.6191278696060181,
      "eval_runtime": 8.6292,
      "eval_samples_per_second": 434.569,
      "eval_steps_per_second": 6.837,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 157.0359649658203,
      "learning_rate": 4.374999999999999e-05,
      "loss": 0.5597,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8648,
      "eval_loss": 0.4498800039291382,
      "eval_runtime": 8.5866,
      "eval_samples_per_second": 436.725,
      "eval_steps_per_second": 6.871,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 354.2825927734375,
      "learning_rate": 3.5e-05,
      "loss": 0.422,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.872,
      "eval_loss": 0.3914497196674347,
      "eval_runtime": 8.8596,
      "eval_samples_per_second": 423.272,
      "eval_steps_per_second": 6.659,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 122.22830963134766,
      "learning_rate": 2.6249999999999998e-05,
      "loss": 0.3414,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8866666666666667,
      "eval_loss": 0.35458439588546753,
      "eval_runtime": 8.7547,
      "eval_samples_per_second": 428.341,
      "eval_steps_per_second": 6.739,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 127.57028198242188,
      "learning_rate": 1.75e-05,
      "loss": 0.2921,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8837333333333334,
      "eval_loss": 0.35498151183128357,
      "eval_runtime": 9.2899,
      "eval_samples_per_second": 403.665,
      "eval_steps_per_second": 6.351,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 101.54278564453125,
      "learning_rate": 8.75e-06,
      "loss": 0.2595,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8845333333333333,
      "eval_loss": 0.34465473890304565,
      "eval_runtime": 8.8893,
      "eval_samples_per_second": 421.854,
      "eval_steps_per_second": 6.637,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 84.95807647705078,
      "learning_rate": 0.0,
      "loss": 0.2392,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.892,
      "eval_loss": 0.3445916771888733,
      "eval_runtime": 8.8112,
      "eval_samples_per_second": 425.595,
      "eval_steps_per_second": 6.696,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}