trainer_state.json

{
  "best_metric": 0.8778666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0253/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 99.77120971679688,
      "learning_rate": 8.75e-05,
      "loss": 2.8882,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7024,
      "eval_loss": 1.1099265813827515,
      "eval_runtime": 8.6097,
      "eval_samples_per_second": 435.554,
      "eval_steps_per_second": 6.853,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 308.15386962890625,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.7913,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8232,
      "eval_loss": 0.5759178400039673,
      "eval_runtime": 8.7605,
      "eval_samples_per_second": 428.055,
      "eval_steps_per_second": 6.735,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 94.36911010742188,
      "learning_rate": 6.25e-05,
      "loss": 0.501,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.844,
      "eval_loss": 0.4824148416519165,
      "eval_runtime": 8.4654,
      "eval_samples_per_second": 442.979,
      "eval_steps_per_second": 6.97,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 97.48247528076172,
      "learning_rate": 5e-05,
      "loss": 0.388,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8570666666666666,
      "eval_loss": 0.4423597455024719,
      "eval_runtime": 8.3357,
      "eval_samples_per_second": 449.87,
      "eval_steps_per_second": 7.078,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 74.7210922241211,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.3104,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8765333333333334,
      "eval_loss": 0.3990243673324585,
      "eval_runtime": 8.5555,
      "eval_samples_per_second": 438.315,
      "eval_steps_per_second": 6.896,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 81.2548599243164,
      "learning_rate": 2.5e-05,
      "loss": 0.2602,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.40288326144218445,
      "eval_runtime": 8.8375,
      "eval_samples_per_second": 424.327,
      "eval_steps_per_second": 6.676,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}