trainer_state.json

{
  "best_metric": 0.8210666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0969/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": 92.3173828125,
      "learning_rate": 2.8514533018536286e-05,
      "loss": 3.8347,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4357333333333333,
      "eval_loss": 3.7008068561553955,
      "eval_runtime": 7.0142,
      "eval_samples_per_second": 534.626,
      "eval_steps_per_second": 8.411,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 208.8642578125,
      "learning_rate": 2.4352347027881003e-05,
      "loss": 2.9696,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6672,
      "eval_loss": 2.1854348182678223,
      "eval_runtime": 6.9385,
      "eval_samples_per_second": 540.464,
      "eval_steps_per_second": 8.503,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 279.7516174316406,
      "learning_rate": 1.8337814009344716e-05,
      "loss": 1.4514,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7584,
      "eval_loss": 1.0753757953643799,
      "eval_runtime": 6.9052,
      "eval_samples_per_second": 543.071,
      "eval_steps_per_second": 8.544,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 81.91796112060547,
      "learning_rate": 1.1662185990655285e-05,
      "loss": 0.9192,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8042666666666667,
      "eval_loss": 0.8227541446685791,
      "eval_runtime": 6.9183,
      "eval_samples_per_second": 542.042,
      "eval_steps_per_second": 8.528,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 283.6077575683594,
      "learning_rate": 5.647652972118998e-06,
      "loss": 0.752,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8168,
      "eval_loss": 0.7604994773864746,
      "eval_runtime": 7.1388,
      "eval_samples_per_second": 525.299,
      "eval_steps_per_second": 8.265,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 328.14373779296875,
      "learning_rate": 1.4854669814637145e-06,
      "loss": 0.6886,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8210666666666666,
      "eval_loss": 0.7058204412460327,
      "eval_runtime": 6.869,
      "eval_samples_per_second": 545.935,
      "eval_steps_per_second": 8.589,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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
}