trainer_state.json

{
  "best_metric": 0.7209302325581395,
  "best_model_checkpoint": "vit-base-patch16-224/checkpoint-12",
  "epoch": 9.6,
  "eval_steps": 500,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.96,
      "eval_accuracy": 0.3488372093023256,
      "eval_loss": 0.742401123046875,
      "eval_runtime": 0.4622,
      "eval_samples_per_second": 93.03,
      "eval_steps_per_second": 23.798,
      "step": 6
    },
    {
      "epoch": 1.6,
      "grad_norm": 6.224728107452393,
      "learning_rate": 4.62962962962963e-06,
      "loss": 0.7374,
      "step": 10
    },
    {
      "epoch": 1.92,
      "eval_accuracy": 0.7209302325581395,
      "eval_loss": 0.5932016372680664,
      "eval_runtime": 0.5117,
      "eval_samples_per_second": 84.03,
      "eval_steps_per_second": 21.496,
      "step": 12
    },
    {
      "epoch": 2.88,
      "eval_accuracy": 0.7209302325581395,
      "eval_loss": 0.5843029618263245,
      "eval_runtime": 0.5094,
      "eval_samples_per_second": 84.414,
      "eval_steps_per_second": 21.594,
      "step": 18
    },
    {
      "epoch": 3.2,
      "grad_norm": 6.278665065765381,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.5783,
      "step": 20
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7209302325581395,
      "eval_loss": 0.599581241607666,
      "eval_runtime": 0.5003,
      "eval_samples_per_second": 85.944,
      "eval_steps_per_second": 21.986,
      "step": 25
    },
    {
      "epoch": 4.8,
      "grad_norm": 12.106355667114258,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 0.5358,
      "step": 30
    },
    {
      "epoch": 4.96,
      "eval_accuracy": 0.7209302325581395,
      "eval_loss": 0.6146759986877441,
      "eval_runtime": 0.5458,
      "eval_samples_per_second": 78.786,
      "eval_steps_per_second": 20.154,
      "step": 31
    },
    {
      "epoch": 5.92,
      "eval_accuracy": 0.7209302325581395,
      "eval_loss": 0.6159453988075256,
      "eval_runtime": 0.4886,
      "eval_samples_per_second": 88.012,
      "eval_steps_per_second": 22.515,
      "step": 37
    },
    {
      "epoch": 6.4,
      "grad_norm": 6.217154502868652,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.5745,
      "step": 40
    },
    {
      "epoch": 6.88,
      "eval_accuracy": 0.7209302325581395,
      "eval_loss": 0.6090837121009827,
      "eval_runtime": 0.4608,
      "eval_samples_per_second": 93.312,
      "eval_steps_per_second": 23.871,
      "step": 43
    },
    {
      "epoch": 8.0,
      "grad_norm": 6.606484413146973,
      "learning_rate": 9.259259259259259e-07,
      "loss": 0.5325,
      "step": 50
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7209302325581395,
      "eval_loss": 0.6066591739654541,
      "eval_runtime": 0.5381,
      "eval_samples_per_second": 79.913,
      "eval_steps_per_second": 20.443,
      "step": 50
    },
    {
      "epoch": 8.96,
      "eval_accuracy": 0.7209302325581395,
      "eval_loss": 0.6046953797340393,
      "eval_runtime": 0.5485,
      "eval_samples_per_second": 78.392,
      "eval_steps_per_second": 20.054,
      "step": 56
    },
    {
      "epoch": 9.6,
      "grad_norm": 4.127104759216309,
      "learning_rate": 0.0,
      "loss": 0.524,
      "step": 60
    },
    {
      "epoch": 9.6,
      "eval_accuracy": 0.7209302325581395,
      "eval_loss": 0.6046017408370972,
      "eval_runtime": 0.7447,
      "eval_samples_per_second": 57.745,
      "eval_steps_per_second": 14.772,
      "step": 60
    },
    {
      "epoch": 9.6,
      "step": 60,
      "total_flos": 9.748371434294477e+16,
      "train_loss": 0.5804096937179566,
      "train_runtime": 57.1953,
      "train_samples_per_second": 17.309,
      "train_steps_per_second": 1.049
    }
  ],
  "logging_steps": 10,
  "max_steps": 60,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 9.748371434294477e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}