| { | |
| "best_metric": 0.753408648227503, | |
| "best_model_checkpoint": "final_models/glue_models/structroberta_s2_final//finetune/qnli/checkpoint-2800", | |
| "epoch": 10.0, | |
| "global_step": 3660, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.09, | |
| "eval_accuracy": 0.6706036925315857, | |
| "eval_f1": 0.6829473684210527, | |
| "eval_loss": 0.6097736358642578, | |
| "eval_mcc": 0.34098528883845747, | |
| "eval_runtime": 4.4598, | |
| "eval_samples_per_second": 512.584, | |
| "eval_steps_per_second": 64.129, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 4.316939890710383e-05, | |
| "loss": 0.6375, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_accuracy": 0.7029746174812317, | |
| "eval_f1": 0.7369236729949632, | |
| "eval_loss": 0.628186821937561, | |
| "eval_mcc": 0.4021712387477629, | |
| "eval_runtime": 4.4598, | |
| "eval_samples_per_second": 512.581, | |
| "eval_steps_per_second": 64.129, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 3.633879781420765e-05, | |
| "loss": 0.51, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "eval_accuracy": 0.7209098935127258, | |
| "eval_f1": 0.7352697095435684, | |
| "eval_loss": 0.6323110461235046, | |
| "eval_mcc": 0.4402833462259992, | |
| "eval_runtime": 4.4581, | |
| "eval_samples_per_second": 512.77, | |
| "eval_steps_per_second": 64.152, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 2.9508196721311478e-05, | |
| "loss": 0.3504, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "eval_accuracy": 0.7139107584953308, | |
| "eval_f1": 0.7240506329113925, | |
| "eval_loss": 0.7396315932273865, | |
| "eval_mcc": 0.4280337957730822, | |
| "eval_runtime": 4.4588, | |
| "eval_samples_per_second": 512.693, | |
| "eval_steps_per_second": 64.143, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 2.2677595628415303e-05, | |
| "loss": 0.201, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "eval_accuracy": 0.7274715900421143, | |
| "eval_f1": 0.7498996386993175, | |
| "eval_loss": 0.9048255085945129, | |
| "eval_mcc": 0.4515715609283487, | |
| "eval_runtime": 4.455, | |
| "eval_samples_per_second": 513.134, | |
| "eval_steps_per_second": 64.198, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "eval_accuracy": 0.7134733200073242, | |
| "eval_f1": 0.7473968376397994, | |
| "eval_loss": 1.0615947246551514, | |
| "eval_mcc": 0.42387486870128077, | |
| "eval_runtime": 4.4578, | |
| "eval_samples_per_second": 512.81, | |
| "eval_steps_per_second": 64.157, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 1.5846994535519128e-05, | |
| "loss": 0.122, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "eval_accuracy": 0.7230970859527588, | |
| "eval_f1": 0.753408648227503, | |
| "eval_loss": 1.2307891845703125, | |
| "eval_mcc": 0.4430099135879824, | |
| "eval_runtime": 4.4561, | |
| "eval_samples_per_second": 513.008, | |
| "eval_steps_per_second": 64.182, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 9.016393442622952e-06, | |
| "loss": 0.0663, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "eval_accuracy": 0.7261592149734497, | |
| "eval_f1": 0.7469684721099435, | |
| "eval_loss": 1.3391324281692505, | |
| "eval_mcc": 0.4491318608834407, | |
| "eval_runtime": 4.456, | |
| "eval_samples_per_second": 513.021, | |
| "eval_steps_per_second": 64.184, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "learning_rate": 2.185792349726776e-06, | |
| "loss": 0.0397, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "eval_accuracy": 0.7178477644920349, | |
| "eval_f1": 0.7376982513216753, | |
| "eval_loss": 1.4820964336395264, | |
| "eval_mcc": 0.4326777561298063, | |
| "eval_runtime": 4.4526, | |
| "eval_samples_per_second": 513.41, | |
| "eval_steps_per_second": 64.232, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 3660, | |
| "total_flos": 3.664831551515136e+16, | |
| "train_loss": 0.26451984220515184, | |
| "train_runtime": 1950.6925, | |
| "train_samples_per_second": 225.135, | |
| "train_steps_per_second": 1.876 | |
| } | |
| ], | |
| "max_steps": 3660, | |
| "num_train_epochs": 10, | |
| "total_flos": 3.664831551515136e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |