{ "best_metric": 0.9, "best_model_checkpoint": "convnextv2-tiny-22k-384-finetuned-zkdeng/t5spiders/checkpoint-18", "epoch": 4.8, "eval_steps": 500, "global_step": 30, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.96, "eval_accuracy": 0.68, "eval_f1": 0.6750754604872251, "eval_loss": 1.1803159713745117, "eval_precision": 0.6789160839160839, "eval_recall": 0.6868831168831169, "eval_runtime": 3.4169, "eval_samples_per_second": 14.633, "eval_steps_per_second": 1.171, "step": 6 }, { "epoch": 1.6, "learning_rate": 3.7037037037037037e-05, "loss": 1.3235, "step": 10 }, { "epoch": 1.92, "eval_accuracy": 0.84, "eval_f1": 0.8387969924812031, "eval_loss": 0.7862005829811096, "eval_precision": 0.8603030303030303, "eval_recall": 0.856969696969697, "eval_runtime": 2.6641, "eval_samples_per_second": 18.768, "eval_steps_per_second": 1.501, "step": 12 }, { "epoch": 2.88, "eval_accuracy": 0.9, "eval_f1": 0.9023978919631095, "eval_loss": 0.5749024152755737, "eval_precision": 0.9034848484848486, "eval_recall": 0.9103030303030304, "eval_runtime": 2.6511, "eval_samples_per_second": 18.86, "eval_steps_per_second": 1.509, "step": 18 }, { "epoch": 3.2, "learning_rate": 1.8518518518518518e-05, "loss": 0.6609, "step": 20 }, { "epoch": 4.0, "eval_accuracy": 0.9, "eval_f1": 0.9023978919631095, "eval_loss": 0.47980961203575134, "eval_precision": 0.9034848484848486, "eval_recall": 0.9103030303030304, "eval_runtime": 2.6994, "eval_samples_per_second": 18.523, "eval_steps_per_second": 1.482, "step": 25 }, { "epoch": 4.8, "learning_rate": 0.0, "loss": 0.43, "step": 30 }, { "epoch": 4.8, "eval_accuracy": 0.9, "eval_f1": 0.9023978919631095, "eval_loss": 0.4612491726875305, "eval_precision": 0.9034848484848486, "eval_recall": 0.9103030303030304, "eval_runtime": 2.6906, "eval_samples_per_second": 18.583, "eval_steps_per_second": 1.487, "step": 30 }, { "epoch": 4.8, "step": 30, "total_flos": 1.4114178534172262e+17, "train_loss": 0.8048010031382243, "train_runtime": 101.5046, "train_samples_per_second": 19.556, "train_steps_per_second": 0.296 } ], "logging_steps": 10, "max_steps": 30, "num_train_epochs": 5, "save_steps": 500, "total_flos": 1.4114178534172262e+17, "trial_name": null, "trial_params": null }