trainer_state.json

{
  "best_metric": 0.7546666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0366/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": 99.27471923828125,
      "learning_rate": 2.8514533018536286e-05,
      "loss": 3.8484,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3181333333333333,
      "eval_loss": 3.7167303562164307,
      "eval_runtime": 13.7792,
      "eval_samples_per_second": 272.15,
      "eval_steps_per_second": 4.282,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 80.9676284790039,
      "learning_rate": 2.4352347027881003e-05,
      "loss": 3.2926,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.52,
      "eval_loss": 2.72161602973938,
      "eval_runtime": 13.1053,
      "eval_samples_per_second": 286.144,
      "eval_steps_per_second": 4.502,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 84.51323699951172,
      "learning_rate": 1.8337814009344716e-05,
      "loss": 2.0957,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6528,
      "eval_loss": 1.6279293298721313,
      "eval_runtime": 12.9792,
      "eval_samples_per_second": 288.924,
      "eval_steps_per_second": 4.546,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 245.26829528808594,
      "learning_rate": 1.1662185990655285e-05,
      "loss": 1.443,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7213333333333334,
      "eval_loss": 1.2210882902145386,
      "eval_runtime": 12.9537,
      "eval_samples_per_second": 289.493,
      "eval_steps_per_second": 4.555,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 173.25588989257812,
      "learning_rate": 5.647652972118998e-06,
      "loss": 1.2006,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.748,
      "eval_loss": 1.1186244487762451,
      "eval_runtime": 12.9151,
      "eval_samples_per_second": 290.358,
      "eval_steps_per_second": 4.568,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 109.20288848876953,
      "learning_rate": 1.4854669814637145e-06,
      "loss": 1.0904,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7546666666666667,
      "eval_loss": 1.0557732582092285,
      "eval_runtime": 13.3488,
      "eval_samples_per_second": 280.923,
      "eval_steps_per_second": 4.42,
      "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
}