trainer_state.json

{
  "best_metric": 0.6589333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0644/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": 63.34593200683594,
      "learning_rate": 2.7135254915624213e-05,
      "loss": 3.856,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3626666666666667,
      "eval_loss": 3.7381067276000977,
      "eval_runtime": 13.4487,
      "eval_samples_per_second": 278.838,
      "eval_steps_per_second": 4.387,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 236.12969970703125,
      "learning_rate": 1.963525491562421e-05,
      "loss": 3.3786,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5269333333333334,
      "eval_loss": 2.956608295440674,
      "eval_runtime": 13.2384,
      "eval_samples_per_second": 283.266,
      "eval_steps_per_second": 4.457,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 300.8673095703125,
      "learning_rate": 1.036474508437579e-05,
      "loss": 2.3336,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6189333333333333,
      "eval_loss": 1.9785981178283691,
      "eval_runtime": 13.4781,
      "eval_samples_per_second": 278.229,
      "eval_steps_per_second": 4.377,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 234.5963134765625,
      "learning_rate": 2.86474508437579e-06,
      "loss": 1.7698,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6357333333333334,
      "eval_loss": 1.7079380750656128,
      "eval_runtime": 13.0204,
      "eval_samples_per_second": 288.01,
      "eval_steps_per_second": 4.531,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 191.47171020507812,
      "learning_rate": 0.0,
      "loss": 1.6416,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6589333333333334,
      "eval_loss": 1.709689736366272,
      "eval_runtime": 13.1851,
      "eval_samples_per_second": 284.412,
      "eval_steps_per_second": 4.475,
      "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
}