{ "best_metric": 0.8105263157894737, "best_model_checkpoint": "vit-msn-small-finetuned-eurosat/checkpoint-58", "epoch": 5.617021276595745, "eval_steps": 500, "global_step": 66, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.851063829787234, "grad_norm": 16.84075164794922, "learning_rate": 0.0001898305084745763, "loss": 1.115, "step": 10 }, { "epoch": 0.9361702127659575, "eval_accuracy": 0.6526315789473685, "eval_loss": 1.039728045463562, "eval_runtime": 5.6197, "eval_samples_per_second": 33.81, "eval_steps_per_second": 1.068, "step": 11 }, { "epoch": 1.702127659574468, "grad_norm": 11.648308753967285, "learning_rate": 0.00015593220338983051, "loss": 0.8536, "step": 20 }, { "epoch": 1.9574468085106385, "eval_accuracy": 0.7578947368421053, "eval_loss": 0.7698105573654175, "eval_runtime": 6.4598, "eval_samples_per_second": 29.413, "eval_steps_per_second": 0.929, "step": 23 }, { "epoch": 2.5531914893617023, "grad_norm": 10.502930641174316, "learning_rate": 0.00012203389830508477, "loss": 0.5677, "step": 30 }, { "epoch": 2.978723404255319, "eval_accuracy": 0.7894736842105263, "eval_loss": 0.7199588418006897, "eval_runtime": 6.391, "eval_samples_per_second": 29.729, "eval_steps_per_second": 0.939, "step": 35 }, { "epoch": 3.404255319148936, "grad_norm": 11.857769012451172, "learning_rate": 8.813559322033899e-05, "loss": 0.419, "step": 40 }, { "epoch": 4.0, "eval_accuracy": 0.7842105263157895, "eval_loss": 0.7286169528961182, "eval_runtime": 6.4892, "eval_samples_per_second": 29.279, "eval_steps_per_second": 0.925, "step": 47 }, { "epoch": 4.25531914893617, "grad_norm": 8.081114768981934, "learning_rate": 5.423728813559322e-05, "loss": 0.3365, "step": 50 }, { "epoch": 4.9361702127659575, "eval_accuracy": 0.8105263157894737, "eval_loss": 0.6607011556625366, "eval_runtime": 5.5742, "eval_samples_per_second": 34.086, "eval_steps_per_second": 1.076, "step": 58 }, { "epoch": 5.1063829787234045, "grad_norm": 6.424224376678467, "learning_rate": 2.033898305084746e-05, "loss": 0.2317, "step": 60 }, { "epoch": 5.617021276595745, "eval_accuracy": 0.8, "eval_loss": 0.6648939251899719, "eval_runtime": 6.1483, "eval_samples_per_second": 30.903, "eval_steps_per_second": 0.976, "step": 66 }, { "epoch": 5.617021276595745, "step": 66, "total_flos": 1.6494617153632666e+17, "train_loss": 0.5514849839788495, "train_runtime": 364.4981, "train_samples_per_second": 24.691, "train_steps_per_second": 0.181 } ], "logging_steps": 10, "max_steps": 66, "num_input_tokens_seen": 0, "num_train_epochs": 6, "save_steps": 500, "total_flos": 1.6494617153632666e+17, "train_batch_size": 32, "trial_name": null, "trial_params": null }