trainer_state.json

{
  "best_metric": 0.7906666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0104/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 149.39297485351562,
      "learning_rate": 2.9095389311788626e-05,
      "loss": 3.8699,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.29413333333333336,
      "eval_loss": 3.8035221099853516,
      "eval_runtime": 5.689,
      "eval_samples_per_second": 659.165,
      "eval_steps_per_second": 10.371,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 77.50113677978516,
      "learning_rate": 2.649066664678467e-05,
      "loss": 3.3249,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5298666666666667,
      "eval_loss": 2.7163941860198975,
      "eval_runtime": 5.4906,
      "eval_samples_per_second": 682.983,
      "eval_steps_per_second": 10.746,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 108.15937042236328,
      "learning_rate": 2.25e-05,
      "loss": 1.892,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6733333333333333,
      "eval_loss": 1.356057047843933,
      "eval_runtime": 5.5759,
      "eval_samples_per_second": 672.536,
      "eval_steps_per_second": 10.581,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 156.11643981933594,
      "learning_rate": 1.760472266500396e-05,
      "loss": 1.2142,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7466666666666667,
      "eval_loss": 0.9784178137779236,
      "eval_runtime": 5.5789,
      "eval_samples_per_second": 672.18,
      "eval_steps_per_second": 10.576,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 223.24295043945312,
      "learning_rate": 1.2395277334996045e-05,
      "loss": 0.9663,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7589333333333333,
      "eval_loss": 0.8321682214736938,
      "eval_runtime": 5.5234,
      "eval_samples_per_second": 678.931,
      "eval_steps_per_second": 10.682,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 222.62301635742188,
      "learning_rate": 7.500000000000004e-06,
      "loss": 0.8553,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7792,
      "eval_loss": 0.7678864598274231,
      "eval_runtime": 5.5106,
      "eval_samples_per_second": 680.503,
      "eval_steps_per_second": 10.707,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 120.20071411132812,
      "learning_rate": 3.5093333532153316e-06,
      "loss": 0.8077,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7818666666666667,
      "eval_loss": 0.7423941493034363,
      "eval_runtime": 5.5079,
      "eval_samples_per_second": 680.838,
      "eval_steps_per_second": 10.712,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 166.0465850830078,
      "learning_rate": 9.046106882113753e-07,
      "loss": 0.7842,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7906666666666666,
      "eval_loss": 0.71159428358078,
      "eval_runtime": 5.461,
      "eval_samples_per_second": 686.684,
      "eval_steps_per_second": 10.804,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}