trainer_state.json

{
  "best_metric": 0.8797333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0479/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": 100.21288299560547,
      "learning_rate": 0.0003,
      "loss": 1.5793,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8197333333333333,
      "eval_loss": 0.5854962468147278,
      "eval_runtime": 15.3523,
      "eval_samples_per_second": 244.263,
      "eval_steps_per_second": 3.843,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 45.63652038574219,
      "learning_rate": 0.0003,
      "loss": 0.4961,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8626666666666667,
      "eval_loss": 0.4550464451313019,
      "eval_runtime": 15.3078,
      "eval_samples_per_second": 244.972,
      "eval_steps_per_second": 3.854,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 84.48673248291016,
      "learning_rate": 0.0003,
      "loss": 0.3249,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8685333333333334,
      "eval_loss": 0.4368308484554291,
      "eval_runtime": 15.4253,
      "eval_samples_per_second": 243.108,
      "eval_steps_per_second": 3.825,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 50.851783752441406,
      "learning_rate": 0.0003,
      "loss": 0.2204,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8738666666666667,
      "eval_loss": 0.45318466424942017,
      "eval_runtime": 15.3411,
      "eval_samples_per_second": 244.441,
      "eval_steps_per_second": 3.846,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 30.559829711914062,
      "learning_rate": 0.0003,
      "loss": 0.1765,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8794666666666666,
      "eval_loss": 0.42279258370399475,
      "eval_runtime": 15.3072,
      "eval_samples_per_second": 244.983,
      "eval_steps_per_second": 3.854,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 55.90869140625,
      "learning_rate": 0.0003,
      "loss": 0.1336,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8797333333333334,
      "eval_loss": 0.4408658742904663,
      "eval_runtime": 15.3038,
      "eval_samples_per_second": 245.038,
      "eval_steps_per_second": 3.855,
      "step": 1998
    }
  ],
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}