trainer_state.json

{
  "best_metric": 0.82,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0728/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": 41.11787033081055,
      "learning_rate": 2.8514533018536286e-05,
      "loss": 3.8039,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4704,
      "eval_loss": 3.5792036056518555,
      "eval_runtime": 15.7818,
      "eval_samples_per_second": 237.615,
      "eval_steps_per_second": 3.738,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 51.831947326660156,
      "learning_rate": 2.4352347027881003e-05,
      "loss": 3.0081,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6656,
      "eval_loss": 2.4305477142333984,
      "eval_runtime": 14.7739,
      "eval_samples_per_second": 253.826,
      "eval_steps_per_second": 3.994,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 197.91024780273438,
      "learning_rate": 1.8337814009344716e-05,
      "loss": 1.6533,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7474666666666666,
      "eval_loss": 1.1312402486801147,
      "eval_runtime": 14.4495,
      "eval_samples_per_second": 259.524,
      "eval_steps_per_second": 4.083,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 137.80955505371094,
      "learning_rate": 1.1662185990655285e-05,
      "loss": 0.9948,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7890666666666667,
      "eval_loss": 0.861015260219574,
      "eval_runtime": 14.3409,
      "eval_samples_per_second": 261.489,
      "eval_steps_per_second": 4.114,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 148.0216064453125,
      "learning_rate": 5.647652972118998e-06,
      "loss": 0.822,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8074666666666667,
      "eval_loss": 0.7683064937591553,
      "eval_runtime": 14.3635,
      "eval_samples_per_second": 261.079,
      "eval_steps_per_second": 4.108,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 237.52947998046875,
      "learning_rate": 1.4854669814637145e-06,
      "loss": 0.7559,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.82,
      "eval_loss": 0.7580312490463257,
      "eval_runtime": 14.3592,
      "eval_samples_per_second": 261.157,
      "eval_steps_per_second": 4.109,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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
}