trainer_state.json

{
  "best_metric": 0.9016,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0245/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": 101.6607666015625,
      "learning_rate": 9.504844339512095e-05,
      "loss": 2.647,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.772,
      "eval_loss": 0.8004442453384399,
      "eval_runtime": 7.8669,
      "eval_samples_per_second": 476.679,
      "eval_steps_per_second": 7.5,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 197.28057861328125,
      "learning_rate": 8.117449009293668e-05,
      "loss": 0.6238,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8597333333333333,
      "eval_loss": 0.46096667647361755,
      "eval_runtime": 7.6517,
      "eval_samples_per_second": 490.085,
      "eval_steps_per_second": 7.711,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 206.0350341796875,
      "learning_rate": 6.112604669781572e-05,
      "loss": 0.3809,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8802666666666666,
      "eval_loss": 0.36226174235343933,
      "eval_runtime": 7.7363,
      "eval_samples_per_second": 484.727,
      "eval_steps_per_second": 7.626,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 117.3563003540039,
      "learning_rate": 3.887395330218429e-05,
      "loss": 0.2822,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8946666666666667,
      "eval_loss": 0.32895806431770325,
      "eval_runtime": 7.6696,
      "eval_samples_per_second": 488.94,
      "eval_steps_per_second": 7.693,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 85.04855346679688,
      "learning_rate": 1.8825509907063327e-05,
      "loss": 0.2141,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8973333333333333,
      "eval_loss": 0.32591986656188965,
      "eval_runtime": 7.6499,
      "eval_samples_per_second": 490.201,
      "eval_steps_per_second": 7.712,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 104.0528793334961,
      "learning_rate": 4.951556604879048e-06,
      "loss": 0.1833,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9016,
      "eval_loss": 0.3201809227466583,
      "eval_runtime": 7.7762,
      "eval_samples_per_second": 482.243,
      "eval_steps_per_second": 7.587,
      "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
}