trainer_state.json

{
  "best_metric": 0.8712,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0336/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": 176.41171264648438,
      "learning_rate": 9e-05,
      "loss": 2.9318,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6930666666666667,
      "eval_loss": 1.1963779926300049,
      "eval_runtime": 9.6976,
      "eval_samples_per_second": 386.696,
      "eval_steps_per_second": 6.084,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 81.95234680175781,
      "learning_rate": 9e-05,
      "loss": 0.8434,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8229333333333333,
      "eval_loss": 0.6208588480949402,
      "eval_runtime": 9.1152,
      "eval_samples_per_second": 411.399,
      "eval_steps_per_second": 6.473,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 75.82111358642578,
      "learning_rate": 9e-05,
      "loss": 0.5116,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8578666666666667,
      "eval_loss": 0.47960788011550903,
      "eval_runtime": 8.9849,
      "eval_samples_per_second": 417.368,
      "eval_steps_per_second": 6.567,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 102.115966796875,
      "learning_rate": 9e-05,
      "loss": 0.3729,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8624,
      "eval_loss": 0.45345059037208557,
      "eval_runtime": 9.1523,
      "eval_samples_per_second": 409.732,
      "eval_steps_per_second": 6.446,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 88.021240234375,
      "learning_rate": 9e-05,
      "loss": 0.2823,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8712,
      "eval_loss": 0.4353785216808319,
      "eval_runtime": 9.2266,
      "eval_samples_per_second": 406.432,
      "eval_steps_per_second": 6.395,
      "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
}