| { | |
| "best_metric": 0.824, | |
| "best_model_checkpoint": "Bert_tinybert-distilled/run-5/checkpoint-1032", | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 1376, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.9512056112289429, | |
| "learning_rate": 0.0001860785294707651, | |
| "loss": 0.5131, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_AUC": 0.835936, | |
| "eval_BAC": 0.7566666666666666, | |
| "eval_MCC": 0.513333789630238, | |
| "eval_PR_AUC": 0.7961733871717276, | |
| "eval_accuracy": 0.7566666666666667, | |
| "eval_loss": 0.45322349667549133, | |
| "eval_runtime": 6.3408, | |
| "eval_samples_per_second": 236.562, | |
| "eval_steps_per_second": 7.412, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.6722192764282227, | |
| "learning_rate": 0.00015949588240351295, | |
| "loss": 0.4704, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_AUC": 0.8548942222222222, | |
| "eval_BAC": 0.7866666666666666, | |
| "eval_MCC": 0.596741247380097, | |
| "eval_PR_AUC": 0.8018353018281653, | |
| "eval_accuracy": 0.7866666666666666, | |
| "eval_loss": 0.43870097398757935, | |
| "eval_runtime": 6.2585, | |
| "eval_samples_per_second": 239.673, | |
| "eval_steps_per_second": 7.51, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.272857904434204, | |
| "learning_rate": 0.0001329132353362608, | |
| "loss": 0.4584, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_AUC": 0.8628124444444446, | |
| "eval_BAC": 0.7953333333333333, | |
| "eval_MCC": 0.5909950885250521, | |
| "eval_PR_AUC": 0.8297629946290537, | |
| "eval_accuracy": 0.7953333333333333, | |
| "eval_loss": 0.42132893204689026, | |
| "eval_runtime": 6.2287, | |
| "eval_samples_per_second": 240.819, | |
| "eval_steps_per_second": 7.546, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.723982810974121, | |
| "learning_rate": 0.00010633058826900864, | |
| "loss": 0.4493, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_AUC": 0.8751004444444445, | |
| "eval_BAC": 0.8133333333333334, | |
| "eval_MCC": 0.629739455225116, | |
| "eval_PR_AUC": 0.842213368644533, | |
| "eval_accuracy": 0.8133333333333334, | |
| "eval_loss": 0.40822649002075195, | |
| "eval_runtime": 6.284, | |
| "eval_samples_per_second": 238.7, | |
| "eval_steps_per_second": 7.479, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.5007638931274414, | |
| "learning_rate": 7.974794120175647e-05, | |
| "loss": 0.44, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_AUC": 0.8788462222222223, | |
| "eval_BAC": 0.8160000000000001, | |
| "eval_MCC": 0.6433072129217033, | |
| "eval_PR_AUC": 0.8432956799918935, | |
| "eval_accuracy": 0.816, | |
| "eval_loss": 0.4106001853942871, | |
| "eval_runtime": 6.2449, | |
| "eval_samples_per_second": 240.196, | |
| "eval_steps_per_second": 7.526, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.927262306213379, | |
| "learning_rate": 5.316529413450432e-05, | |
| "loss": 0.434, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_AUC": 0.8820462222222223, | |
| "eval_BAC": 0.8240000000000001, | |
| "eval_MCC": 0.6523025718935186, | |
| "eval_PR_AUC": 0.8496646661140822, | |
| "eval_accuracy": 0.824, | |
| "eval_loss": 0.40657976269721985, | |
| "eval_runtime": 6.2506, | |
| "eval_samples_per_second": 239.975, | |
| "eval_steps_per_second": 7.519, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 3.092992067337036, | |
| "learning_rate": 2.658264706725216e-05, | |
| "loss": 0.4302, | |
| "step": 1204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_AUC": 0.8827448888888889, | |
| "eval_BAC": 0.8240000000000001, | |
| "eval_MCC": 0.6521028966567438, | |
| "eval_PR_AUC": 0.8533179574845196, | |
| "eval_accuracy": 0.824, | |
| "eval_loss": 0.40630367398262024, | |
| "eval_runtime": 6.2493, | |
| "eval_samples_per_second": 240.026, | |
| "eval_steps_per_second": 7.521, | |
| "step": 1204 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.997439980506897, | |
| "learning_rate": 0.0, | |
| "loss": 0.4247, | |
| "step": 1376 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_AUC": 0.8827804444444444, | |
| "eval_BAC": 0.8213333333333332, | |
| "eval_MCC": 0.6461706581381526, | |
| "eval_PR_AUC": 0.8514881992132012, | |
| "eval_accuracy": 0.8213333333333334, | |
| "eval_loss": 0.4052167236804962, | |
| "eval_runtime": 6.27, | |
| "eval_samples_per_second": 239.234, | |
| "eval_steps_per_second": 7.496, | |
| "step": 1376 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1376, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "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": 3603021840000.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.7927861374990359, | |
| "learning_rate": 0.00021266117653801728, | |
| "num_train_epochs": 8, | |
| "temperature": 5 | |
| } | |
| } | |