trainer_state.json

{
  "best_metric": 0.868,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0156/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 26.107681274414062,
      "learning_rate": 0.0005,
      "loss": 1.3997,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8072,
      "eval_loss": 0.6410058736801147,
      "eval_runtime": 13.4225,
      "eval_samples_per_second": 279.381,
      "eval_steps_per_second": 4.396,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 32.148529052734375,
      "learning_rate": 0.0005,
      "loss": 0.5182,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8544,
      "eval_loss": 0.4776754677295685,
      "eval_runtime": 13.62,
      "eval_samples_per_second": 275.33,
      "eval_steps_per_second": 4.332,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 36.779869079589844,
      "learning_rate": 0.0005,
      "loss": 0.3514,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8546666666666667,
      "eval_loss": 0.4985320568084717,
      "eval_runtime": 13.441,
      "eval_samples_per_second": 278.997,
      "eval_steps_per_second": 4.39,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 22.98525619506836,
      "learning_rate": 0.0005,
      "loss": 0.2641,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8613333333333333,
      "eval_loss": 0.5044768452644348,
      "eval_runtime": 13.9169,
      "eval_samples_per_second": 269.456,
      "eval_steps_per_second": 4.239,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 67.30899810791016,
      "learning_rate": 0.0005,
      "loss": 0.2171,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.868,
      "eval_loss": 0.46776342391967773,
      "eval_runtime": 13.3647,
      "eval_samples_per_second": 280.59,
      "eval_steps_per_second": 4.415,
      "step": 1665
    }
  ],
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}