| { | |
| "best_metric": 5.599513085818624, | |
| "best_model_checkpoint": "./Whisper-squeezeformer-v3\\checkpoint-40000", | |
| "epoch": 20.0, | |
| "eval_steps": 2500, | |
| "global_step": 50000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 10.493926048278809, | |
| "learning_rate": 9.972e-06, | |
| "loss": 4.8718, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.8609094619750977, | |
| "eval_runtime": 650.8319, | |
| "eval_samples_per_second": 4.026, | |
| "eval_steps_per_second": 0.504, | |
| "eval_wer": 111.85902312842362, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 6.234286785125732, | |
| "learning_rate": 8.575428571428573e-06, | |
| "loss": 2.5628, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.2978440523147583, | |
| "eval_runtime": 624.1365, | |
| "eval_samples_per_second": 4.198, | |
| "eval_steps_per_second": 0.526, | |
| "eval_wer": 15.619293974437007, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.2328641414642334, | |
| "learning_rate": 7.146857142857143e-06, | |
| "loss": 0.1698, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.22175125777721405, | |
| "eval_runtime": 585.8606, | |
| "eval_samples_per_second": 4.472, | |
| "eval_steps_per_second": 0.56, | |
| "eval_wer": 11.090611685940353, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.424901247024536, | |
| "learning_rate": 5.718285714285715e-06, | |
| "loss": 0.0867, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.2011481672525406, | |
| "eval_runtime": 614.9021, | |
| "eval_samples_per_second": 4.261, | |
| "eval_steps_per_second": 0.533, | |
| "eval_wer": 10.189059646987218, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.77018404006958, | |
| "learning_rate": 4.290285714285714e-06, | |
| "loss": 0.1697, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.1640562266111374, | |
| "eval_runtime": 597.6823, | |
| "eval_samples_per_second": 4.384, | |
| "eval_steps_per_second": 0.549, | |
| "eval_wer": 8.985088253195373, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.6757357120513916, | |
| "learning_rate": 2.861714285714286e-06, | |
| "loss": 0.0993, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.15530936419963837, | |
| "eval_runtime": 585.6998, | |
| "eval_samples_per_second": 4.473, | |
| "eval_steps_per_second": 0.56, | |
| "eval_wer": 7.803940961655509, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.374549388885498, | |
| "learning_rate": 1.434857142857143e-06, | |
| "loss": 0.0651, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.1555396020412445, | |
| "eval_runtime": 585.2066, | |
| "eval_samples_per_second": 4.477, | |
| "eval_steps_per_second": 0.56, | |
| "eval_wer": 7.244750456482045, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.977560043334961, | |
| "learning_rate": 6.2857142857142855e-09, | |
| "loss": 0.0468, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.15691713988780975, | |
| "eval_runtime": 580.4295, | |
| "eval_samples_per_second": 4.514, | |
| "eval_steps_per_second": 0.565, | |
| "eval_wer": 7.149650030432136, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 4.3119354248046875, | |
| "learning_rate": 2.731636363636364e-06, | |
| "loss": 0.2168, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.15093587338924408, | |
| "eval_runtime": 583.1378, | |
| "eval_samples_per_second": 4.493, | |
| "eval_steps_per_second": 0.562, | |
| "eval_wer": 7.050745587340232, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.5476253032684326, | |
| "learning_rate": 1.822909090909091e-06, | |
| "loss": 0.1467, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.1494235247373581, | |
| "eval_runtime": 591.6025, | |
| "eval_samples_per_second": 4.429, | |
| "eval_steps_per_second": 0.554, | |
| "eval_wer": 6.967057212416311, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 3.3021419048309326, | |
| "learning_rate": 9.141818181818182e-07, | |
| "loss": 0.1113, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.14934100210666656, | |
| "eval_runtime": 584.5502, | |
| "eval_samples_per_second": 4.482, | |
| "eval_steps_per_second": 0.561, | |
| "eval_wer": 6.75973828362751, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 2.4910900592803955, | |
| "learning_rate": 5.090909090909091e-09, | |
| "loss": 0.0914, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.15113578736782074, | |
| "eval_runtime": 619.6762, | |
| "eval_samples_per_second": 4.228, | |
| "eval_steps_per_second": 0.529, | |
| "eval_wer": 6.80348447961047, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 4.114705562591553, | |
| "learning_rate": 2.004e-06, | |
| "loss": 0.1946, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.139073446393013, | |
| "eval_runtime": 595.1979, | |
| "eval_samples_per_second": 4.402, | |
| "eval_steps_per_second": 0.551, | |
| "eval_wer": 6.421180766889837, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 4.109494686126709, | |
| "learning_rate": 1.3378666666666667e-06, | |
| "loss": 0.1425, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.13689054548740387, | |
| "eval_runtime": 633.0271, | |
| "eval_samples_per_second": 4.139, | |
| "eval_steps_per_second": 0.518, | |
| "eval_wer": 5.87530432136336, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 2.726956367492676, | |
| "learning_rate": 6.712e-07, | |
| "loss": 0.1145, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.13682714104652405, | |
| "eval_runtime": 601.9865, | |
| "eval_samples_per_second": 4.352, | |
| "eval_steps_per_second": 0.545, | |
| "eval_wer": 5.753575776019477, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 3.426710844039917, | |
| "learning_rate": 1.1807058823529414e-06, | |
| "loss": 0.1776, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.13023081421852112, | |
| "eval_runtime": 578.9166, | |
| "eval_samples_per_second": 4.526, | |
| "eval_steps_per_second": 0.567, | |
| "eval_wer": 5.599513085818624, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 3.3029887676239014, | |
| "learning_rate": 5.927058823529412e-07, | |
| "loss": 0.1416, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.12984110414981842, | |
| "eval_runtime": 624.3393, | |
| "eval_samples_per_second": 4.196, | |
| "eval_steps_per_second": 0.525, | |
| "eval_wer": 5.620435179549604, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.106375217437744, | |
| "learning_rate": 4.705882352941177e-09, | |
| "loss": 0.1239, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.12974976003170013, | |
| "eval_runtime": 580.9928, | |
| "eval_samples_per_second": 4.51, | |
| "eval_steps_per_second": 0.565, | |
| "eval_wer": 5.620435179549604, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.747838973999023, | |
| "learning_rate": 5.309473684210527e-07, | |
| "loss": 0.3373, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.13534972071647644, | |
| "eval_runtime": 649.9396, | |
| "eval_samples_per_second": 4.031, | |
| "eval_steps_per_second": 0.505, | |
| "eval_wer": 5.740261716372489, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 5.997547626495361, | |
| "learning_rate": 4.842105263157895e-09, | |
| "loss": 0.2785, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.13221722841262817, | |
| "eval_runtime": 634.989, | |
| "eval_samples_per_second": 4.126, | |
| "eval_steps_per_second": 0.517, | |
| "eval_wer": 5.6641813755325625, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 50000, | |
| "total_flos": 3.4664620032e+20, | |
| "train_loss": 0.030787386474609376, | |
| "train_runtime": 24493.4093, | |
| "train_samples_per_second": 40.827, | |
| "train_steps_per_second": 2.041 | |
| } | |
| ], | |
| "logging_steps": 2500, | |
| "max_steps": 50000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 2500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.4664620032e+20, | |
| "train_batch_size": 20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |