trainer_state.json

{
  "best_metric": 0.9221333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0874/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": 44.865413665771484,
      "learning_rate": 0.0002625,
      "loss": 1.4864,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8642666666666666,
      "eval_loss": 0.46916502714157104,
      "eval_runtime": 7.0037,
      "eval_samples_per_second": 535.432,
      "eval_steps_per_second": 8.424,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 71.48839569091797,
      "learning_rate": 0.000225,
      "loss": 0.3683,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9056,
      "eval_loss": 0.3261888027191162,
      "eval_runtime": 6.9849,
      "eval_samples_per_second": 536.871,
      "eval_steps_per_second": 8.447,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 43.82332229614258,
      "learning_rate": 0.00018749999999999998,
      "loss": 0.188,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9077333333333333,
      "eval_loss": 0.3181098401546478,
      "eval_runtime": 6.9333,
      "eval_samples_per_second": 540.871,
      "eval_steps_per_second": 8.51,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 25.575328826904297,
      "learning_rate": 0.00015,
      "loss": 0.1072,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9168,
      "eval_loss": 0.2922641336917877,
      "eval_runtime": 6.9457,
      "eval_samples_per_second": 539.903,
      "eval_steps_per_second": 8.494,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 47.711700439453125,
      "learning_rate": 0.0001125,
      "loss": 0.059,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9168,
      "eval_loss": 0.31680163741111755,
      "eval_runtime": 7.2094,
      "eval_samples_per_second": 520.154,
      "eval_steps_per_second": 8.184,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 59.391937255859375,
      "learning_rate": 7.5e-05,
      "loss": 0.0347,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9221333333333334,
      "eval_loss": 0.3129202723503113,
      "eval_runtime": 6.9251,
      "eval_samples_per_second": 541.512,
      "eval_steps_per_second": 8.52,
      "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
}