trainer_state.json

{
  "best_metric": 0.7898666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0794/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": 168.471435546875,
      "learning_rate": 2.9095389311788626e-05,
      "loss": 3.8457,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.34613333333333335,
      "eval_loss": 3.7308104038238525,
      "eval_runtime": 7.7271,
      "eval_samples_per_second": 485.306,
      "eval_steps_per_second": 7.635,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 119.71832275390625,
      "learning_rate": 2.649066664678467e-05,
      "loss": 3.2159,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5322666666666667,
      "eval_loss": 2.7048842906951904,
      "eval_runtime": 7.7051,
      "eval_samples_per_second": 486.694,
      "eval_steps_per_second": 7.657,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 128.01878356933594,
      "learning_rate": 2.25e-05,
      "loss": 1.9269,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6890666666666667,
      "eval_loss": 1.4137322902679443,
      "eval_runtime": 7.8466,
      "eval_samples_per_second": 477.917,
      "eval_steps_per_second": 7.519,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 264.3545837402344,
      "learning_rate": 1.760472266500396e-05,
      "loss": 1.2283,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7450666666666667,
      "eval_loss": 1.0189623832702637,
      "eval_runtime": 7.82,
      "eval_samples_per_second": 479.538,
      "eval_steps_per_second": 7.545,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 115.18975830078125,
      "learning_rate": 1.2395277334996045e-05,
      "loss": 0.9511,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7642666666666666,
      "eval_loss": 0.8596847653388977,
      "eval_runtime": 7.7152,
      "eval_samples_per_second": 486.051,
      "eval_steps_per_second": 7.647,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 312.8894958496094,
      "learning_rate": 7.500000000000004e-06,
      "loss": 0.821,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7848,
      "eval_loss": 0.7762792706489563,
      "eval_runtime": 7.8378,
      "eval_samples_per_second": 478.453,
      "eval_steps_per_second": 7.528,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 191.66091918945312,
      "learning_rate": 3.5093333532153316e-06,
      "loss": 0.752,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7837333333333333,
      "eval_loss": 0.7603239417076111,
      "eval_runtime": 7.8028,
      "eval_samples_per_second": 480.596,
      "eval_steps_per_second": 7.561,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 291.5173034667969,
      "learning_rate": 9.046106882113753e-07,
      "loss": 0.7359,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7898666666666667,
      "eval_loss": 0.7383450269699097,
      "eval_runtime": 7.7952,
      "eval_samples_per_second": 481.063,
      "eval_steps_per_second": 7.569,
      "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
}