trainer_state.json

{
  "best_metric": 0.8944,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0215/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 155.11871337890625,
      "learning_rate": 0.0001,
      "loss": 2.6819,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7541333333333333,
      "eval_loss": 0.8786283731460571,
      "eval_runtime": 12.7329,
      "eval_samples_per_second": 294.512,
      "eval_steps_per_second": 4.634,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 145.08253479003906,
      "learning_rate": 0.0001,
      "loss": 0.6451,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8485333333333334,
      "eval_loss": 0.4883054792881012,
      "eval_runtime": 12.8017,
      "eval_samples_per_second": 292.93,
      "eval_steps_per_second": 4.609,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 147.68507385253906,
      "learning_rate": 0.0001,
      "loss": 0.4049,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.868,
      "eval_loss": 0.43600741028785706,
      "eval_runtime": 12.6555,
      "eval_samples_per_second": 296.314,
      "eval_steps_per_second": 4.662,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 76.3962173461914,
      "learning_rate": 0.0001,
      "loss": 0.2778,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8861333333333333,
      "eval_loss": 0.38121041655540466,
      "eval_runtime": 12.6395,
      "eval_samples_per_second": 296.69,
      "eval_steps_per_second": 4.668,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 66.27699279785156,
      "learning_rate": 0.0001,
      "loss": 0.2015,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.888,
      "eval_loss": 0.36607205867767334,
      "eval_runtime": 12.7012,
      "eval_samples_per_second": 295.248,
      "eval_steps_per_second": 4.645,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 64.09796905517578,
      "learning_rate": 0.0001,
      "loss": 0.1546,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.888,
      "eval_loss": 0.3744758665561676,
      "eval_runtime": 12.6576,
      "eval_samples_per_second": 296.265,
      "eval_steps_per_second": 4.661,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 55.5106201171875,
      "learning_rate": 0.0001,
      "loss": 0.1224,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8944,
      "eval_loss": 0.3898276686668396,
      "eval_runtime": 12.7477,
      "eval_samples_per_second": 294.171,
      "eval_steps_per_second": 4.628,
      "step": 2331
    }
  ],
  "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": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}