trainer_state.json

{
  "best_metric": 0.9050666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0458/checkpoints/checkpoint-2997",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 2997,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 162.13372802734375,
      "learning_rate": 8e-05,
      "loss": 2.8654,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7645333333333333,
      "eval_loss": 0.9676640629768372,
      "eval_runtime": 13.5162,
      "eval_samples_per_second": 277.445,
      "eval_steps_per_second": 4.365,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 80.60504913330078,
      "learning_rate": 7.000000000000001e-05,
      "loss": 0.6851,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8610666666666666,
      "eval_loss": 0.4846009314060211,
      "eval_runtime": 13.6528,
      "eval_samples_per_second": 274.67,
      "eval_steps_per_second": 4.321,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 64.14983367919922,
      "learning_rate": 6e-05,
      "loss": 0.4156,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8824,
      "eval_loss": 0.3966742157936096,
      "eval_runtime": 13.4463,
      "eval_samples_per_second": 278.888,
      "eval_steps_per_second": 4.388,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 54.500587463378906,
      "learning_rate": 5e-05,
      "loss": 0.2955,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8928,
      "eval_loss": 0.3779648244380951,
      "eval_runtime": 13.5315,
      "eval_samples_per_second": 277.132,
      "eval_steps_per_second": 4.36,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 90.90096282958984,
      "learning_rate": 4e-05,
      "loss": 0.2315,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8933333333333333,
      "eval_loss": 0.36070793867111206,
      "eval_runtime": 13.5866,
      "eval_samples_per_second": 276.007,
      "eval_steps_per_second": 4.343,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 167.26478576660156,
      "learning_rate": 3e-05,
      "loss": 0.1814,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8992,
      "eval_loss": 0.34440454840660095,
      "eval_runtime": 13.4837,
      "eval_samples_per_second": 278.113,
      "eval_steps_per_second": 4.376,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 88.83964538574219,
      "learning_rate": 2e-05,
      "loss": 0.1569,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9008,
      "eval_loss": 0.3417004644870758,
      "eval_runtime": 13.6275,
      "eval_samples_per_second": 275.179,
      "eval_steps_per_second": 4.329,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 88.92556762695312,
      "learning_rate": 1e-05,
      "loss": 0.137,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9048,
      "eval_loss": 0.3350420594215393,
      "eval_runtime": 15.3201,
      "eval_samples_per_second": 244.777,
      "eval_steps_per_second": 3.851,
      "step": 2664
    },
    {
      "epoch": 9.0,
      "grad_norm": 67.48157501220703,
      "learning_rate": 0.0,
      "loss": 0.1228,
      "step": 2997
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9050666666666667,
      "eval_loss": 0.3314981758594513,
      "eval_runtime": 13.4025,
      "eval_samples_per_second": 279.798,
      "eval_steps_per_second": 4.402,
      "step": 2997
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.3587863155968e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}