| { | |
| "best_metric": 0.21690590111642744, | |
| "best_model_checkpoint": "/content/drive/MyDrive/Graduation project/dataset/part 3/Models/asc_wav2vec2_large_xlsr_model/checkpoint-500", | |
| "epoch": 3.0, | |
| "eval_steps": 100, | |
| "global_step": 1500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.21812936663627625, | |
| "learning_rate": 0.00023, | |
| "loss": 4.0688, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "eval_cer": 1.0, | |
| "eval_loss": 2.944187879562378, | |
| "eval_runtime": 25.2241, | |
| "eval_samples_per_second": 3.964, | |
| "eval_steps_per_second": 3.964, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.5214684009552002, | |
| "learning_rate": 0.00023, | |
| "loss": 1.9289, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "eval_cer": 0.09955781857162, | |
| "eval_loss": 0.5113587379455566, | |
| "eval_runtime": 24.8081, | |
| "eval_samples_per_second": 4.031, | |
| "eval_steps_per_second": 4.031, | |
| "eval_wer": 0.4920255183413078, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "grad_norm": 0.45686057209968567, | |
| "learning_rate": 0.00023, | |
| "loss": 0.3881, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_cer": 0.053463754522310065, | |
| "eval_loss": 0.23284225165843964, | |
| "eval_runtime": 24.918, | |
| "eval_samples_per_second": 4.013, | |
| "eval_steps_per_second": 4.013, | |
| "eval_wer": 0.2711323763955343, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.142857142857142, | |
| "grad_norm": 0.3496246039867401, | |
| "learning_rate": 0.00023, | |
| "loss": 0.1989, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.142857142857142, | |
| "eval_cer": 0.05011389521640091, | |
| "eval_loss": 0.20896148681640625, | |
| "eval_runtime": 25.0682, | |
| "eval_samples_per_second": 3.989, | |
| "eval_steps_per_second": 3.989, | |
| "eval_wer": 0.2567783094098884, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 1.072981357574463, | |
| "learning_rate": 0.00023, | |
| "loss": 0.1418, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "eval_cer": 0.041136272276564384, | |
| "eval_loss": 0.1919952929019928, | |
| "eval_runtime": 24.835, | |
| "eval_samples_per_second": 4.027, | |
| "eval_steps_per_second": 4.027, | |
| "eval_wer": 0.21690590111642744, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 2.501166820526123, | |
| "learning_rate": 0.00023, | |
| "loss": 0.9766, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_cer": 0.22892712711125207, | |
| "eval_loss": 0.7451229095458984, | |
| "eval_runtime": 219.3212, | |
| "eval_samples_per_second": 9.119, | |
| "eval_steps_per_second": 9.119, | |
| "eval_wer": 0.7126639264043845, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 2.21364688873291, | |
| "learning_rate": 0.00023, | |
| "loss": 0.7679, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_cer": 0.20032114820394895, | |
| "eval_loss": 0.6842666864395142, | |
| "eval_runtime": 220.1309, | |
| "eval_samples_per_second": 9.086, | |
| "eval_steps_per_second": 9.086, | |
| "eval_wer": 0.6376981796829125, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 1.8162668943405151, | |
| "learning_rate": 0.00023, | |
| "loss": 0.6767, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_cer": 0.19177702006185077, | |
| "eval_loss": 0.6032456159591675, | |
| "eval_runtime": 218.5745, | |
| "eval_samples_per_second": 9.15, | |
| "eval_steps_per_second": 9.15, | |
| "eval_wer": 0.6313368565276962, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 1.9325016736984253, | |
| "learning_rate": 0.00023, | |
| "loss": 0.6254, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_cer": 0.18921972880818333, | |
| "eval_loss": 0.5750653147697449, | |
| "eval_runtime": 221.3428, | |
| "eval_samples_per_second": 9.036, | |
| "eval_steps_per_second": 9.036, | |
| "eval_wer": 0.6109806224310042, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.9160747528076172, | |
| "learning_rate": 0.00023, | |
| "loss": 0.6156, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_cer": 0.1742526365871065, | |
| "eval_loss": 0.5430945158004761, | |
| "eval_runtime": 219.1129, | |
| "eval_samples_per_second": 9.128, | |
| "eval_steps_per_second": 9.128, | |
| "eval_wer": 0.5729105500097866, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 1.2983200550079346, | |
| "learning_rate": 0.00023, | |
| "loss": 0.4506, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "eval_cer": 0.15419078582190152, | |
| "eval_loss": 0.5200591087341309, | |
| "eval_runtime": 220.7565, | |
| "eval_samples_per_second": 9.06, | |
| "eval_steps_per_second": 9.06, | |
| "eval_wer": 0.5123311802701116, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.5487444400787354, | |
| "learning_rate": 0.00023, | |
| "loss": 0.4399, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_cer": 0.15298152406629134, | |
| "eval_loss": 0.522709310054779, | |
| "eval_runtime": 215.7214, | |
| "eval_samples_per_second": 9.271, | |
| "eval_steps_per_second": 9.271, | |
| "eval_wer": 0.5041103934233705, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 1.8711397647857666, | |
| "learning_rate": 0.00023, | |
| "loss": 0.4346, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "eval_cer": 0.153219411624772, | |
| "eval_loss": 0.5107786059379578, | |
| "eval_runtime": 219.26, | |
| "eval_samples_per_second": 9.122, | |
| "eval_steps_per_second": 9.122, | |
| "eval_wer": 0.5042082599334508, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 1.684812068939209, | |
| "learning_rate": 0.00023, | |
| "loss": 0.4148, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_cer": 0.14915549916739354, | |
| "eval_loss": 0.5022586584091187, | |
| "eval_runtime": 220.0102, | |
| "eval_samples_per_second": 9.09, | |
| "eval_steps_per_second": 9.09, | |
| "eval_wer": 0.4847328244274809, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.5067812204360962, | |
| "learning_rate": 0.0, | |
| "loss": 0.4207, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_cer": 0.1372809452065657, | |
| "eval_loss": 0.4935847222805023, | |
| "eval_runtime": 220.1042, | |
| "eval_samples_per_second": 9.087, | |
| "eval_steps_per_second": 9.087, | |
| "eval_wer": 0.4493051477784302, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 1500, | |
| "total_flos": 4.297170075587716e+18, | |
| "train_loss": 0.38818399047851565, | |
| "train_runtime": 7019.9025, | |
| "train_samples_per_second": 3.419, | |
| "train_steps_per_second": 0.214 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 4.297170075587716e+18, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |