trainer_state.json

{
  "best_metric": 0.8885333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0611/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": 104.02849578857422,
      "learning_rate": 6.788924172750679e-05,
      "loss": 3.3444,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6501333333333333,
      "eval_loss": 1.7333364486694336,
      "eval_runtime": 15.4437,
      "eval_samples_per_second": 242.817,
      "eval_steps_per_second": 3.82,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 184.76019287109375,
      "learning_rate": 6.181155550916422e-05,
      "loss": 1.0477,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8176,
      "eval_loss": 0.6258362531661987,
      "eval_runtime": 15.4811,
      "eval_samples_per_second": 242.231,
      "eval_steps_per_second": 3.811,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 98.88613891601562,
      "learning_rate": 5.2499999999999995e-05,
      "loss": 0.5726,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8565333333333334,
      "eval_loss": 0.488643616437912,
      "eval_runtime": 15.3294,
      "eval_samples_per_second": 244.627,
      "eval_steps_per_second": 3.849,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 68.14766693115234,
      "learning_rate": 4.1077686218342566e-05,
      "loss": 0.4209,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8693333333333333,
      "eval_loss": 0.44844362139701843,
      "eval_runtime": 15.4085,
      "eval_samples_per_second": 243.373,
      "eval_steps_per_second": 3.829,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 118.67496490478516,
      "learning_rate": 2.8922313781657438e-05,
      "loss": 0.3448,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8786666666666667,
      "eval_loss": 0.40992897748947144,
      "eval_runtime": 15.3406,
      "eval_samples_per_second": 244.449,
      "eval_steps_per_second": 3.846,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 65.93782043457031,
      "learning_rate": 1.7500000000000005e-05,
      "loss": 0.2888,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8832,
      "eval_loss": 0.3947669565677643,
      "eval_runtime": 14.4402,
      "eval_samples_per_second": 259.692,
      "eval_steps_per_second": 4.086,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 84.99073028564453,
      "learning_rate": 8.188444490835772e-06,
      "loss": 0.2711,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8821333333333333,
      "eval_loss": 0.4025212228298187,
      "eval_runtime": 14.4029,
      "eval_samples_per_second": 260.364,
      "eval_steps_per_second": 4.096,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 134.485107421875,
      "learning_rate": 2.1107582724932087e-06,
      "loss": 0.251,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8885333333333333,
      "eval_loss": 0.3817089796066284,
      "eval_runtime": 14.1427,
      "eval_samples_per_second": 265.155,
      "eval_steps_per_second": 4.172,
      "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
}