trainer_state.json

{
  "best_metric": 0.8904,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0630/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": 103.55120086669922,
      "learning_rate": 6.788924172750679e-05,
      "loss": 3.1741,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.688,
      "eval_loss": 1.4741727113723755,
      "eval_runtime": 7.9552,
      "eval_samples_per_second": 471.392,
      "eval_steps_per_second": 7.417,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 76.6065902709961,
      "learning_rate": 6.181155550916422e-05,
      "loss": 0.8094,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8330666666666666,
      "eval_loss": 0.5402723550796509,
      "eval_runtime": 7.8173,
      "eval_samples_per_second": 479.707,
      "eval_steps_per_second": 7.547,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 58.45398712158203,
      "learning_rate": 5.2499999999999995e-05,
      "loss": 0.4207,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.41931217908859253,
      "eval_runtime": 7.6303,
      "eval_samples_per_second": 491.464,
      "eval_steps_per_second": 7.732,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 64.42362976074219,
      "learning_rate": 4.1077686218342566e-05,
      "loss": 0.2868,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8765333333333334,
      "eval_loss": 0.38354355096817017,
      "eval_runtime": 7.9826,
      "eval_samples_per_second": 469.774,
      "eval_steps_per_second": 7.391,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 251.53814697265625,
      "learning_rate": 2.8922313781657438e-05,
      "loss": 0.2039,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8856,
      "eval_loss": 0.3581760823726654,
      "eval_runtime": 7.9426,
      "eval_samples_per_second": 472.138,
      "eval_steps_per_second": 7.428,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 162.0258331298828,
      "learning_rate": 1.7500000000000005e-05,
      "loss": 0.16,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8896,
      "eval_loss": 0.3578397333621979,
      "eval_runtime": 7.874,
      "eval_samples_per_second": 476.251,
      "eval_steps_per_second": 7.493,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 124.9599838256836,
      "learning_rate": 8.188444490835772e-06,
      "loss": 0.1304,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8850666666666667,
      "eval_loss": 0.35708171129226685,
      "eval_runtime": 7.8921,
      "eval_samples_per_second": 475.158,
      "eval_steps_per_second": 7.476,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 168.0413360595703,
      "learning_rate": 2.1107582724932087e-06,
      "loss": 0.1192,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8904,
      "eval_loss": 0.349144846200943,
      "eval_runtime": 7.8715,
      "eval_samples_per_second": 476.405,
      "eval_steps_per_second": 7.495,
      "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
}