trainer_state.json

{
  "best_metric": 0.9178666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0396/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": 35.694725036621094,
      "learning_rate": 0.0002851453301853628,
      "loss": 1.5839,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8538666666666667,
      "eval_loss": 0.5192927122116089,
      "eval_runtime": 7.8053,
      "eval_samples_per_second": 480.444,
      "eval_steps_per_second": 7.559,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 116.19212341308594,
      "learning_rate": 0.00024352347027881003,
      "loss": 0.444,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8802666666666666,
      "eval_loss": 0.3959866464138031,
      "eval_runtime": 7.6337,
      "eval_samples_per_second": 491.245,
      "eval_steps_per_second": 7.729,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 30.70604133605957,
      "learning_rate": 0.00018337814009344714,
      "loss": 0.2463,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8941333333333333,
      "eval_loss": 0.3634697198867798,
      "eval_runtime": 7.6844,
      "eval_samples_per_second": 488.0,
      "eval_steps_per_second": 7.678,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 35.00081253051758,
      "learning_rate": 0.00011662185990655284,
      "loss": 0.1467,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8994666666666666,
      "eval_loss": 0.350903183221817,
      "eval_runtime": 7.6813,
      "eval_samples_per_second": 488.198,
      "eval_steps_per_second": 7.681,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 26.199243545532227,
      "learning_rate": 5.6476529721189974e-05,
      "loss": 0.0858,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9112,
      "eval_loss": 0.319852739572525,
      "eval_runtime": 7.7777,
      "eval_samples_per_second": 482.149,
      "eval_steps_per_second": 7.586,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 67.42407989501953,
      "learning_rate": 1.4854669814637143e-05,
      "loss": 0.0599,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9178666666666667,
      "eval_loss": 0.29966607689857483,
      "eval_runtime": 7.758,
      "eval_samples_per_second": 483.371,
      "eval_steps_per_second": 7.605,
      "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
}