trainer_state.json

{
  "best_metric": 0.8754666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0992/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": 169.2122039794922,
      "learning_rate": 9e-05,
      "loss": 2.9542,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6885333333333333,
      "eval_loss": 1.229636549949646,
      "eval_runtime": 15.3922,
      "eval_samples_per_second": 243.63,
      "eval_steps_per_second": 3.833,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 161.28053283691406,
      "learning_rate": 9e-05,
      "loss": 0.8298,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8216,
      "eval_loss": 0.5920099020004272,
      "eval_runtime": 15.2066,
      "eval_samples_per_second": 246.604,
      "eval_steps_per_second": 3.88,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 51.27986145019531,
      "learning_rate": 9e-05,
      "loss": 0.5115,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8565333333333334,
      "eval_loss": 0.4719916880130768,
      "eval_runtime": 15.3257,
      "eval_samples_per_second": 244.687,
      "eval_steps_per_second": 3.85,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 60.64800262451172,
      "learning_rate": 9e-05,
      "loss": 0.3807,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8664,
      "eval_loss": 0.42665135860443115,
      "eval_runtime": 15.0836,
      "eval_samples_per_second": 248.614,
      "eval_steps_per_second": 3.912,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 31.401546478271484,
      "learning_rate": 9e-05,
      "loss": 0.2861,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.43038684129714966,
      "eval_runtime": 14.368,
      "eval_samples_per_second": 260.996,
      "eval_steps_per_second": 4.106,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 128.6389923095703,
      "learning_rate": 9e-05,
      "loss": 0.2292,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8722666666666666,
      "eval_loss": 0.4199257791042328,
      "eval_runtime": 14.4093,
      "eval_samples_per_second": 260.249,
      "eval_steps_per_second": 4.095,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 161.71630859375,
      "learning_rate": 9e-05,
      "loss": 0.1844,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8698666666666667,
      "eval_loss": 0.4446857273578644,
      "eval_runtime": 14.386,
      "eval_samples_per_second": 260.67,
      "eval_steps_per_second": 4.101,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 60.8952522277832,
      "learning_rate": 9e-05,
      "loss": 0.1478,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8754666666666666,
      "eval_loss": 0.43379276990890503,
      "eval_runtime": 14.6725,
      "eval_samples_per_second": 255.58,
      "eval_steps_per_second": 4.021,
      "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
}