{ "best_metric": 0.008335741236805916, "best_model_checkpoint": "convnextv2-tiny-1k-224-finetuned-barkley\\checkpoint-266", "epoch": 10.0, "eval_steps": 500, "global_step": 380, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "train_accuracy": 0.3717105263157895 }, { "epoch": 1.0, "grad_norm": Infinity, "learning_rate": 1.5789473684210526e-05, "loss": 1.4696, "step": 38 }, { "epoch": 1.0, "eval_accuracy": 0.6768426344896932, "eval_error_rate": 0.3231573655103068, "eval_f1": 0.6179742198967869, "eval_loss": 1.1806769371032715, "eval_precision": 0.7272665779298716, "eval_recall": 0.6513157894736842, "eval_runtime": 39.4866, "eval_samples_per_second": 3.849, "eval_steps_per_second": 0.481, "eval_top1_accuracy": 0.6513157894736842, "step": 38 }, { "epoch": 2.0, "train_accuracy": 0.8472222222222222 }, { "epoch": 2.0, "grad_norm": 13.038293838500977, "learning_rate": 3.24561403508772e-05, "loss": 0.7197, "step": 76 }, { "epoch": 2.0, "eval_accuracy": 0.9434188034188035, "eval_error_rate": 0.056581196581196536, "eval_f1": 0.9403973173646983, "eval_loss": 0.3719189167022705, "eval_precision": 0.9439213320792268, "eval_recall": 0.9407894736842105, "eval_runtime": 25.8579, "eval_samples_per_second": 5.878, "eval_steps_per_second": 0.735, "eval_top1_accuracy": 0.9473684210526315, "step": 76 }, { "epoch": 3.0, "train_accuracy": 0.9429824561403509 }, { "epoch": 3.0, "grad_norm": 18.88597297668457, "learning_rate": 4.868421052631579e-05, "loss": 0.2388, "step": 114 }, { "epoch": 3.0, "eval_accuracy": 0.9715686274509803, "eval_error_rate": 0.02843137254901973, "eval_f1": 0.9670619516200449, "eval_loss": 0.14889617264270782, "eval_precision": 0.9687561893444246, "eval_recall": 0.9671052631578947, "eval_runtime": 56.7711, "eval_samples_per_second": 2.677, "eval_steps_per_second": 0.335, "eval_top1_accuracy": 0.9671052631578947, "step": 114 }, { "epoch": 4.0, "train_accuracy": 0.9758771929824561 }, { "epoch": 4.0, "grad_norm": 20.039169311523438, "learning_rate": 4.834307992202729e-05, "loss": 0.1048, "step": 152 }, { "epoch": 4.0, "eval_accuracy": 0.9877777777777779, "eval_error_rate": 0.012222222222222134, "eval_f1": 0.9868421052631579, "eval_loss": 0.07301948219537735, "eval_precision": 0.9868421052631579, "eval_recall": 0.9868421052631579, "eval_runtime": 57.9769, "eval_samples_per_second": 2.622, "eval_steps_per_second": 0.328, "eval_top1_accuracy": 0.9868421052631579, "step": 152 }, { "epoch": 5.0, "train_accuracy": 0.9685672514619883 }, { "epoch": 5.0, "grad_norm": 32.21807098388672, "learning_rate": 4.6539961013645225e-05, "loss": 0.1103, "step": 190 }, { "epoch": 5.0, "eval_accuracy": 0.9877777777777779, "eval_error_rate": 0.012222222222222134, "eval_f1": 0.9868421052631579, "eval_loss": 0.028847167268395424, "eval_precision": 0.9868421052631579, "eval_recall": 0.9868421052631579, "eval_runtime": 55.6388, "eval_samples_per_second": 2.732, "eval_steps_per_second": 0.341, "eval_top1_accuracy": 0.9868421052631579, "step": 190 }, { "epoch": 6.0, "train_accuracy": 0.9824561403508771 }, { "epoch": 6.0, "grad_norm": 0.3600322902202606, "learning_rate": 4.468810916179337e-05, "loss": 0.072, "step": 228 }, { "epoch": 6.0, "eval_accuracy": 0.9867521367521368, "eval_error_rate": 0.0132478632478632, "eval_f1": 0.986943499272542, "eval_loss": 0.053720876574516296, "eval_precision": 0.9876644736842105, "eval_recall": 0.9868421052631579, "eval_runtime": 58.3931, "eval_samples_per_second": 2.603, "eval_steps_per_second": 0.325, "eval_top1_accuracy": 0.9868421052631579, "step": 228 }, { "epoch": 7.0, "train_accuracy": 0.9919590643274854 }, { "epoch": 7.0, "grad_norm": 0.21095441281795502, "learning_rate": 4.283625730994152e-05, "loss": 0.0248, "step": 266 }, { "epoch": 7.0, "eval_accuracy": 1.0, "eval_error_rate": 0.0, "eval_f1": 1.0, "eval_loss": 0.008335741236805916, "eval_precision": 1.0, "eval_recall": 1.0, "eval_runtime": 60.881, "eval_samples_per_second": 2.497, "eval_steps_per_second": 0.312, "eval_top1_accuracy": 1.0, "step": 266 }, { "epoch": 8.0, "train_accuracy": 0.9904970760233918 }, { "epoch": 8.0, "grad_norm": 65.39556121826172, "learning_rate": 4.098440545808967e-05, "loss": 0.0371, "step": 304 }, { "epoch": 8.0, "eval_accuracy": 0.9800000000000001, "eval_error_rate": 0.019999999999999907, "eval_f1": 0.9801607428504544, "eval_loss": 0.06527390331029892, "eval_precision": 0.9818634423897581, "eval_recall": 0.9802631578947368, "eval_runtime": 54.5902, "eval_samples_per_second": 2.784, "eval_steps_per_second": 0.348, "eval_top1_accuracy": 0.9802631578947368, "step": 304 }, { "epoch": 9.0, "train_accuracy": 0.9846491228070176 }, { "epoch": 9.0, "grad_norm": 38.53361511230469, "learning_rate": 3.9181286549707604e-05, "loss": 0.0626, "step": 342 }, { "epoch": 9.0, "eval_accuracy": 0.95, "eval_error_rate": 0.050000000000000044, "eval_f1": 0.9404217064400131, "eval_loss": 0.22711969912052155, "eval_precision": 0.9544534412955467, "eval_recall": 0.9407894736842105, "eval_runtime": 51.6016, "eval_samples_per_second": 2.946, "eval_steps_per_second": 0.368, "eval_top1_accuracy": 0.9407894736842105, "step": 342 }, { "epoch": 10.0, "train_accuracy": 0.9766081871345029 }, { "epoch": 10.0, "grad_norm": 0.2995816767215729, "learning_rate": 3.7329434697855756e-05, "loss": 0.07, "step": 380 }, { "epoch": 10.0, "eval_accuracy": 0.9933333333333334, "eval_error_rate": 0.006666666666666599, "eval_f1": 0.9934103601236665, "eval_loss": 0.030387282371520996, "eval_precision": 0.9935988620199147, "eval_recall": 0.993421052631579, "eval_runtime": 52.2379, "eval_samples_per_second": 2.91, "eval_steps_per_second": 0.364, "eval_top1_accuracy": 0.993421052631579, "step": 380 }, { "epoch": 10.0, "step": 380, "total_flos": 3.060866577059021e+17, "train_loss": 0.2909727611039814, "train_runtime": 5060.0554, "train_samples_per_second": 7.209, "train_steps_per_second": 0.225 } ], "logging_steps": 10, "max_steps": 1140, "num_input_tokens_seen": 0, "num_train_epochs": 30, "save_steps": 500, "stateful_callbacks": { "EarlyStoppingCallback": { "args": { "early_stopping_patience": 3, "early_stopping_threshold": 0.0 }, "attributes": { "early_stopping_patience_counter": 0 } }, "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 3.060866577059021e+17, "train_batch_size": 8, "trial_name": null, "trial_params": null }