trainer_state.json

{
  "best_metric": 0.9106666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0129/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": 41.82192611694336,
      "learning_rate": 0.0002713525491562421,
      "loss": 1.5552,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8517333333333333,
      "eval_loss": 0.5223625898361206,
      "eval_runtime": 15.6693,
      "eval_samples_per_second": 239.321,
      "eval_steps_per_second": 3.765,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 38.85689163208008,
      "learning_rate": 0.0001963525491562421,
      "loss": 0.4325,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8992,
      "eval_loss": 0.35599005222320557,
      "eval_runtime": 15.5864,
      "eval_samples_per_second": 240.594,
      "eval_steps_per_second": 3.785,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 22.43977165222168,
      "learning_rate": 0.0001036474508437579,
      "loss": 0.2488,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9029333333333334,
      "eval_loss": 0.3160007894039154,
      "eval_runtime": 15.4661,
      "eval_samples_per_second": 242.466,
      "eval_steps_per_second": 3.815,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 40.43728256225586,
      "learning_rate": 2.8647450843757897e-05,
      "loss": 0.1571,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9085333333333333,
      "eval_loss": 0.3177623152732849,
      "eval_runtime": 15.57,
      "eval_samples_per_second": 240.849,
      "eval_steps_per_second": 3.789,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 29.620681762695312,
      "learning_rate": 0.0,
      "loss": 0.1164,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9106666666666666,
      "eval_loss": 0.30620408058166504,
      "eval_runtime": 15.5376,
      "eval_samples_per_second": 241.35,
      "eval_steps_per_second": 3.797,
      "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
}