trainer_state.json

{
  "best_metric": 0.9069333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0618/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": 45.330482482910156,
      "learning_rate": 0.0002625,
      "loss": 1.4974,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8485333333333334,
      "eval_loss": 0.5177019834518433,
      "eval_runtime": 7.4719,
      "eval_samples_per_second": 501.881,
      "eval_steps_per_second": 7.896,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 41.72834396362305,
      "learning_rate": 0.000225,
      "loss": 0.348,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8765333333333334,
      "eval_loss": 0.4077262282371521,
      "eval_runtime": 7.4521,
      "eval_samples_per_second": 503.216,
      "eval_steps_per_second": 7.917,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 30.190841674804688,
      "learning_rate": 0.00018749999999999998,
      "loss": 0.1753,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8904,
      "eval_loss": 0.3836580514907837,
      "eval_runtime": 7.3988,
      "eval_samples_per_second": 506.837,
      "eval_steps_per_second": 7.974,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 38.53179931640625,
      "learning_rate": 0.00015,
      "loss": 0.0967,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8965333333333333,
      "eval_loss": 0.381554514169693,
      "eval_runtime": 7.3954,
      "eval_samples_per_second": 507.069,
      "eval_steps_per_second": 7.978,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 32.856651306152344,
      "learning_rate": 0.0001125,
      "loss": 0.0464,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8962666666666667,
      "eval_loss": 0.42138612270355225,
      "eval_runtime": 7.407,
      "eval_samples_per_second": 506.277,
      "eval_steps_per_second": 7.965,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 77.25407409667969,
      "learning_rate": 7.5e-05,
      "loss": 0.0313,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9069333333333334,
      "eval_loss": 0.40026426315307617,
      "eval_runtime": 7.3566,
      "eval_samples_per_second": 509.744,
      "eval_steps_per_second": 8.02,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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
}