trainer_state.json

{
  "best_metric": 0.6424,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0089/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": 546.5914916992188,
      "learning_rate": 2.7135254915624213e-05,
      "loss": 3.8499,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.33093333333333336,
      "eval_loss": 3.741302967071533,
      "eval_runtime": 8.3625,
      "eval_samples_per_second": 448.431,
      "eval_steps_per_second": 7.055,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 197.605712890625,
      "learning_rate": 1.963525491562421e-05,
      "loss": 3.3375,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5506666666666666,
      "eval_loss": 2.918170213699341,
      "eval_runtime": 8.1083,
      "eval_samples_per_second": 462.491,
      "eval_steps_per_second": 7.277,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 275.83673095703125,
      "learning_rate": 1.036474508437579e-05,
      "loss": 2.4312,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6061333333333333,
      "eval_loss": 2.064338207244873,
      "eval_runtime": 8.3086,
      "eval_samples_per_second": 451.339,
      "eval_steps_per_second": 7.101,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 244.60501098632812,
      "learning_rate": 2.86474508437579e-06,
      "loss": 1.8811,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6354666666666666,
      "eval_loss": 1.7561602592468262,
      "eval_runtime": 7.8284,
      "eval_samples_per_second": 479.023,
      "eval_steps_per_second": 7.537,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 185.083740234375,
      "learning_rate": 0.0,
      "loss": 1.7343,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6424,
      "eval_loss": 1.6944528818130493,
      "eval_runtime": 7.5077,
      "eval_samples_per_second": 499.485,
      "eval_steps_per_second": 7.859,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}