trainer_state.json

{
  "best_metric": 0.8429333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0474/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 71.62551879882812,
      "learning_rate": 4.522542485937369e-05,
      "loss": 3.6957,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5314666666666666,
      "eval_loss": 3.1336379051208496,
      "eval_runtime": 7.2772,
      "eval_samples_per_second": 515.306,
      "eval_steps_per_second": 8.107,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 72.27831268310547,
      "learning_rate": 3.272542485937369e-05,
      "loss": 1.8729,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7570666666666667,
      "eval_loss": 0.9842764139175415,
      "eval_runtime": 7.351,
      "eval_samples_per_second": 510.132,
      "eval_steps_per_second": 8.026,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 330.38897705078125,
      "learning_rate": 1.7274575140626318e-05,
      "loss": 0.805,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8176,
      "eval_loss": 0.6920135617256165,
      "eval_runtime": 7.2664,
      "eval_samples_per_second": 516.074,
      "eval_steps_per_second": 8.12,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 113.3585433959961,
      "learning_rate": 4.7745751406263165e-06,
      "loss": 0.5968,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8288,
      "eval_loss": 0.5907721519470215,
      "eval_runtime": 8.0149,
      "eval_samples_per_second": 467.876,
      "eval_steps_per_second": 7.361,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 207.1141815185547,
      "learning_rate": 0.0,
      "loss": 0.5553,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8429333333333333,
      "eval_loss": 0.5761522650718689,
      "eval_runtime": 7.6962,
      "eval_samples_per_second": 487.252,
      "eval_steps_per_second": 7.666,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}