trainer_state.json

{
  "best_metric": 0.8780487804878049,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-60",
  "epoch": 2.962962962962963,
  "eval_steps": 500,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.49,
      "grad_norm": 7.887508392333984,
      "learning_rate": 4.62962962962963e-05,
      "loss": 1.1976,
      "step": 10
    },
    {
      "epoch": 0.99,
      "grad_norm": 9.747188568115234,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.7533,
      "step": 20
    },
    {
      "epoch": 0.99,
      "eval_accuracy": 0.7770034843205574,
      "eval_loss": 0.5473685264587402,
      "eval_runtime": 4.0122,
      "eval_samples_per_second": 71.531,
      "eval_steps_per_second": 2.243,
      "step": 20
    },
    {
      "epoch": 1.48,
      "grad_norm": 18.254377365112305,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.5298,
      "step": 30
    },
    {
      "epoch": 1.98,
      "grad_norm": 10.268461227416992,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.4366,
      "step": 40
    },
    {
      "epoch": 1.98,
      "eval_accuracy": 0.8606271777003485,
      "eval_loss": 0.3385283052921295,
      "eval_runtime": 3.9279,
      "eval_samples_per_second": 73.066,
      "eval_steps_per_second": 2.291,
      "step": 40
    },
    {
      "epoch": 2.47,
      "grad_norm": 11.050724983215332,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.3945,
      "step": 50
    },
    {
      "epoch": 2.96,
      "grad_norm": 5.141947269439697,
      "learning_rate": 0.0,
      "loss": 0.3318,
      "step": 60
    },
    {
      "epoch": 2.96,
      "eval_accuracy": 0.8780487804878049,
      "eval_loss": 0.28312796354293823,
      "eval_runtime": 5.5586,
      "eval_samples_per_second": 51.632,
      "eval_steps_per_second": 1.619,
      "step": 60
    },
    {
      "epoch": 2.96,
      "step": 60,
      "total_flos": 1.9045723068223488e+17,
      "train_loss": 0.6072771787643433,
      "train_runtime": 158.0756,
      "train_samples_per_second": 49.021,
      "train_steps_per_second": 0.38
    }
  ],
  "logging_steps": 10,
  "max_steps": 60,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1.9045723068223488e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}