trainer_state.json

{
  "best_metric": 0.9093333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0483/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": 122.4942626953125,
      "learning_rate": 0.0002909538931178862,
      "loss": 1.6372,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8290666666666666,
      "eval_loss": 0.5949521064758301,
      "eval_runtime": 7.7118,
      "eval_samples_per_second": 486.266,
      "eval_steps_per_second": 7.651,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 50.07439041137695,
      "learning_rate": 0.00026490666646784665,
      "loss": 0.482,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8653333333333333,
      "eval_loss": 0.46260717511177063,
      "eval_runtime": 7.6239,
      "eval_samples_per_second": 491.874,
      "eval_steps_per_second": 7.739,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 61.31623840332031,
      "learning_rate": 0.000225,
      "loss": 0.3085,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8808,
      "eval_loss": 0.397551029920578,
      "eval_runtime": 7.5704,
      "eval_samples_per_second": 495.353,
      "eval_steps_per_second": 7.794,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 37.782527923583984,
      "learning_rate": 0.00017604722665003956,
      "loss": 0.204,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8925333333333333,
      "eval_loss": 0.3769581913948059,
      "eval_runtime": 7.5745,
      "eval_samples_per_second": 495.081,
      "eval_steps_per_second": 7.789,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 33.443115234375,
      "learning_rate": 0.00012395277334996044,
      "loss": 0.1318,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8922666666666667,
      "eval_loss": 0.3887059688568115,
      "eval_runtime": 7.602,
      "eval_samples_per_second": 493.293,
      "eval_steps_per_second": 7.761,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 53.05466842651367,
      "learning_rate": 7.500000000000002e-05,
      "loss": 0.0968,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9021333333333333,
      "eval_loss": 0.35605087876319885,
      "eval_runtime": 7.5641,
      "eval_samples_per_second": 495.764,
      "eval_steps_per_second": 7.8,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 34.67047119140625,
      "learning_rate": 3.509333353215331e-05,
      "loss": 0.0586,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9058666666666667,
      "eval_loss": 0.3699686825275421,
      "eval_runtime": 7.6992,
      "eval_samples_per_second": 487.063,
      "eval_steps_per_second": 7.663,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 48.235626220703125,
      "learning_rate": 9.046106882113751e-06,
      "loss": 0.0472,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9093333333333333,
      "eval_loss": 0.3523571193218231,
      "eval_runtime": 8.0272,
      "eval_samples_per_second": 467.161,
      "eval_steps_per_second": 7.35,
      "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
}