trainer_state.json

{
  "best_metric": 0.8874666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0263/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": 115.40930938720703,
      "learning_rate": 0.0001,
      "loss": 2.8325,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7082666666666667,
      "eval_loss": 1.0947363376617432,
      "eval_runtime": 5.46,
      "eval_samples_per_second": 686.809,
      "eval_steps_per_second": 10.806,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 57.42362976074219,
      "learning_rate": 0.0001,
      "loss": 0.782,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8213333333333334,
      "eval_loss": 0.5797080993652344,
      "eval_runtime": 5.4549,
      "eval_samples_per_second": 687.452,
      "eval_steps_per_second": 10.816,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 71.19671630859375,
      "learning_rate": 0.0001,
      "loss": 0.4865,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8458666666666667,
      "eval_loss": 0.5015274286270142,
      "eval_runtime": 5.6024,
      "eval_samples_per_second": 669.356,
      "eval_steps_per_second": 10.531,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 115.84815979003906,
      "learning_rate": 0.0001,
      "loss": 0.3464,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8594666666666667,
      "eval_loss": 0.4603502154350281,
      "eval_runtime": 5.4649,
      "eval_samples_per_second": 686.201,
      "eval_steps_per_second": 10.796,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 54.55414581298828,
      "learning_rate": 0.0001,
      "loss": 0.2502,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8701333333333333,
      "eval_loss": 0.42953911423683167,
      "eval_runtime": 5.4787,
      "eval_samples_per_second": 684.471,
      "eval_steps_per_second": 10.769,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 56.43640899658203,
      "learning_rate": 0.0001,
      "loss": 0.1948,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8837333333333334,
      "eval_loss": 0.3902447521686554,
      "eval_runtime": 5.4181,
      "eval_samples_per_second": 692.124,
      "eval_steps_per_second": 10.889,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 96.23876190185547,
      "learning_rate": 0.0001,
      "loss": 0.1546,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8794666666666666,
      "eval_loss": 0.4360315203666687,
      "eval_runtime": 5.367,
      "eval_samples_per_second": 698.719,
      "eval_steps_per_second": 10.993,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 68.26825714111328,
      "learning_rate": 0.0001,
      "loss": 0.1217,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8874666666666666,
      "eval_loss": 0.41860124468803406,
      "eval_runtime": 5.3642,
      "eval_samples_per_second": 699.081,
      "eval_steps_per_second": 10.999,
      "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
}