| { |
| "best_metric": 0.19209351050339085, |
| "best_model_checkpoint": "./en-xlsr/checkpoint-6000", |
| "epoch": 27.906976744186046, |
| "eval_steps": 600, |
| "global_step": 6000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.47, |
| "learning_rate": 0.0002983555207517619, |
| "loss": 4.591, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.93, |
| "learning_rate": 0.0002936570086139389, |
| "loss": 2.9761, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.4, |
| "learning_rate": 0.0002889584964761159, |
| "loss": 2.8963, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.86, |
| "learning_rate": 0.00028425998433829287, |
| "loss": 2.0367, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.33, |
| "learning_rate": 0.0002795614722004698, |
| "loss": 0.9943, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.79, |
| "learning_rate": 0.0002748629600626468, |
| "loss": 0.8037, |
| "step": 600 |
| }, |
| { |
| "epoch": 2.79, |
| "eval_cer": 0.13958760039582999, |
| "eval_loss": 0.5361498594284058, |
| "eval_runtime": 23.0079, |
| "eval_samples_per_second": 75.539, |
| "eval_steps_per_second": 4.738, |
| "eval_wer": 0.3517671059160831, |
| "step": 600 |
| }, |
| { |
| "epoch": 3.26, |
| "learning_rate": 0.0002701644479248238, |
| "loss": 0.685, |
| "step": 700 |
| }, |
| { |
| "epoch": 3.72, |
| "learning_rate": 0.0002654659357870008, |
| "loss": 0.6178, |
| "step": 800 |
| }, |
| { |
| "epoch": 4.19, |
| "learning_rate": 0.00026076742364917777, |
| "loss": 0.5696, |
| "step": 900 |
| }, |
| { |
| "epoch": 4.65, |
| "learning_rate": 0.0002560689115113547, |
| "loss": 0.5172, |
| "step": 1000 |
| }, |
| { |
| "epoch": 5.12, |
| "learning_rate": 0.0002513703993735317, |
| "loss": 0.4947, |
| "step": 1100 |
| }, |
| { |
| "epoch": 5.58, |
| "learning_rate": 0.0002466718872357087, |
| "loss": 0.4529, |
| "step": 1200 |
| }, |
| { |
| "epoch": 5.58, |
| "eval_cer": 0.10734586794927853, |
| "eval_loss": 0.4010535180568695, |
| "eval_runtime": 22.5531, |
| "eval_samples_per_second": 77.062, |
| "eval_steps_per_second": 4.833, |
| "eval_wer": 0.256657661134697, |
| "step": 1200 |
| }, |
| { |
| "epoch": 6.05, |
| "learning_rate": 0.00024197337509788565, |
| "loss": 0.4454, |
| "step": 1300 |
| }, |
| { |
| "epoch": 6.51, |
| "learning_rate": 0.00023727486296006264, |
| "loss": 0.4102, |
| "step": 1400 |
| }, |
| { |
| "epoch": 6.98, |
| "learning_rate": 0.0002325763508222396, |
| "loss": 0.4059, |
| "step": 1500 |
| }, |
| { |
| "epoch": 7.44, |
| "learning_rate": 0.0002278778386844166, |
| "loss": 0.3731, |
| "step": 1600 |
| }, |
| { |
| "epoch": 7.91, |
| "learning_rate": 0.00022317932654659358, |
| "loss": 0.3747, |
| "step": 1700 |
| }, |
| { |
| "epoch": 8.37, |
| "learning_rate": 0.00021848081440877052, |
| "loss": 0.3482, |
| "step": 1800 |
| }, |
| { |
| "epoch": 8.37, |
| "eval_cer": 0.1060571191862472, |
| "eval_loss": 0.4073833227157593, |
| "eval_runtime": 22.9802, |
| "eval_samples_per_second": 75.63, |
| "eval_steps_per_second": 4.743, |
| "eval_wer": 0.2514748855929867, |
| "step": 1800 |
| }, |
| { |
| "epoch": 8.84, |
| "learning_rate": 0.0002137823022709475, |
| "loss": 0.3388, |
| "step": 1900 |
| }, |
| { |
| "epoch": 9.3, |
| "learning_rate": 0.00020908379013312447, |
| "loss": 0.3296, |
| "step": 2000 |
| }, |
| { |
| "epoch": 9.77, |
| "learning_rate": 0.00020438527799530146, |
| "loss": 0.3247, |
| "step": 2100 |
| }, |
| { |
| "epoch": 10.23, |
| "learning_rate": 0.00019968676585747845, |
| "loss": 0.3057, |
| "step": 2200 |
| }, |
| { |
| "epoch": 10.7, |
| "learning_rate": 0.00019498825371965542, |
| "loss": 0.3022, |
| "step": 2300 |
| }, |
| { |
| "epoch": 11.16, |
| "learning_rate": 0.0001902897415818324, |
| "loss": 0.291, |
| "step": 2400 |
| }, |
| { |
| "epoch": 11.16, |
| "eval_cer": 0.09982049570800636, |
| "eval_loss": 0.3988887071609497, |
| "eval_runtime": 22.5626, |
| "eval_samples_per_second": 77.03, |
| "eval_steps_per_second": 4.831, |
| "eval_wer": 0.22886916248552683, |
| "step": 2400 |
| }, |
| { |
| "epoch": 11.63, |
| "learning_rate": 0.0001855912294440094, |
| "loss": 0.2773, |
| "step": 2500 |
| }, |
| { |
| "epoch": 12.09, |
| "learning_rate": 0.00018089271730618636, |
| "loss": 0.2759, |
| "step": 2600 |
| }, |
| { |
| "epoch": 12.56, |
| "learning_rate": 0.00017619420516836332, |
| "loss": 0.2551, |
| "step": 2700 |
| }, |
| { |
| "epoch": 13.02, |
| "learning_rate": 0.00017149569303054031, |
| "loss": 0.2587, |
| "step": 2800 |
| }, |
| { |
| "epoch": 13.49, |
| "learning_rate": 0.00016679718089271728, |
| "loss": 0.2426, |
| "step": 2900 |
| }, |
| { |
| "epoch": 13.95, |
| "learning_rate": 0.00016209866875489427, |
| "loss": 0.2469, |
| "step": 3000 |
| }, |
| { |
| "epoch": 13.95, |
| "eval_cer": 0.09549398200354406, |
| "eval_loss": 0.40880683064460754, |
| "eval_runtime": 23.2843, |
| "eval_samples_per_second": 74.642, |
| "eval_steps_per_second": 4.681, |
| "eval_wer": 0.21712521365165133, |
| "step": 3000 |
| }, |
| { |
| "epoch": 14.42, |
| "learning_rate": 0.00015740015661707126, |
| "loss": 0.2398, |
| "step": 3100 |
| }, |
| { |
| "epoch": 14.88, |
| "learning_rate": 0.00015270164447924822, |
| "loss": 0.2298, |
| "step": 3200 |
| }, |
| { |
| "epoch": 15.35, |
| "learning_rate": 0.00014800313234142518, |
| "loss": 0.2263, |
| "step": 3300 |
| }, |
| { |
| "epoch": 15.81, |
| "learning_rate": 0.00014330462020360218, |
| "loss": 0.2193, |
| "step": 3400 |
| }, |
| { |
| "epoch": 16.28, |
| "learning_rate": 0.00013860610806577917, |
| "loss": 0.2139, |
| "step": 3500 |
| }, |
| { |
| "epoch": 16.74, |
| "learning_rate": 0.00013390759592795613, |
| "loss": 0.2088, |
| "step": 3600 |
| }, |
| { |
| "epoch": 16.74, |
| "eval_cer": 0.09318113824131821, |
| "eval_loss": 0.4487144649028778, |
| "eval_runtime": 22.9016, |
| "eval_samples_per_second": 75.89, |
| "eval_steps_per_second": 4.759, |
| "eval_wer": 0.20620830346804875, |
| "step": 3600 |
| }, |
| { |
| "epoch": 17.21, |
| "learning_rate": 0.00012920908379013312, |
| "loss": 0.212, |
| "step": 3700 |
| }, |
| { |
| "epoch": 17.67, |
| "learning_rate": 0.00012451057165231008, |
| "loss": 0.1886, |
| "step": 3800 |
| }, |
| { |
| "epoch": 18.14, |
| "learning_rate": 0.00011981205951448707, |
| "loss": 0.1854, |
| "step": 3900 |
| }, |
| { |
| "epoch": 18.6, |
| "learning_rate": 0.00011511354737666405, |
| "loss": 0.1891, |
| "step": 4000 |
| }, |
| { |
| "epoch": 19.07, |
| "learning_rate": 0.00011041503523884103, |
| "loss": 0.1832, |
| "step": 4100 |
| }, |
| { |
| "epoch": 19.53, |
| "learning_rate": 0.000105716523101018, |
| "loss": 0.1828, |
| "step": 4200 |
| }, |
| { |
| "epoch": 19.53, |
| "eval_cer": 0.09201896301751318, |
| "eval_loss": 0.4508346617221832, |
| "eval_runtime": 22.6393, |
| "eval_samples_per_second": 76.769, |
| "eval_steps_per_second": 4.815, |
| "eval_wer": 0.20146661520648398, |
| "step": 4200 |
| }, |
| { |
| "epoch": 20.0, |
| "learning_rate": 0.00010101801096319498, |
| "loss": 0.1805, |
| "step": 4300 |
| }, |
| { |
| "epoch": 20.47, |
| "learning_rate": 9.631949882537196e-05, |
| "loss": 0.1767, |
| "step": 4400 |
| }, |
| { |
| "epoch": 20.93, |
| "learning_rate": 9.162098668754895e-05, |
| "loss": 0.1717, |
| "step": 4500 |
| }, |
| { |
| "epoch": 21.4, |
| "learning_rate": 8.692247454972591e-05, |
| "loss": 0.1592, |
| "step": 4600 |
| }, |
| { |
| "epoch": 21.86, |
| "learning_rate": 8.222396241190289e-05, |
| "loss": 0.1621, |
| "step": 4700 |
| }, |
| { |
| "epoch": 22.33, |
| "learning_rate": 7.752545027407988e-05, |
| "loss": 0.1566, |
| "step": 4800 |
| }, |
| { |
| "epoch": 22.33, |
| "eval_cer": 0.09145513543368697, |
| "eval_loss": 0.4802495539188385, |
| "eval_runtime": 22.7447, |
| "eval_samples_per_second": 76.413, |
| "eval_steps_per_second": 4.792, |
| "eval_wer": 0.19837900424546506, |
| "step": 4800 |
| }, |
| { |
| "epoch": 22.79, |
| "learning_rate": 7.282693813625684e-05, |
| "loss": 0.1531, |
| "step": 4900 |
| }, |
| { |
| "epoch": 23.26, |
| "learning_rate": 6.812842599843383e-05, |
| "loss": 0.1495, |
| "step": 5000 |
| }, |
| { |
| "epoch": 23.72, |
| "learning_rate": 6.34299138606108e-05, |
| "loss": 0.1502, |
| "step": 5100 |
| }, |
| { |
| "epoch": 24.19, |
| "learning_rate": 5.873140172278778e-05, |
| "loss": 0.1412, |
| "step": 5200 |
| }, |
| { |
| "epoch": 24.65, |
| "learning_rate": 5.4032889584964754e-05, |
| "loss": 0.1378, |
| "step": 5300 |
| }, |
| { |
| "epoch": 25.12, |
| "learning_rate": 4.933437744714174e-05, |
| "loss": 0.1414, |
| "step": 5400 |
| }, |
| { |
| "epoch": 25.12, |
| "eval_cer": 0.08983269279451361, |
| "eval_loss": 0.47997498512268066, |
| "eval_runtime": 23.0232, |
| "eval_samples_per_second": 75.489, |
| "eval_steps_per_second": 4.734, |
| "eval_wer": 0.19882009152561064, |
| "step": 5400 |
| }, |
| { |
| "epoch": 25.58, |
| "learning_rate": 4.463586530931871e-05, |
| "loss": 0.1365, |
| "step": 5500 |
| }, |
| { |
| "epoch": 26.05, |
| "learning_rate": 3.993735317149569e-05, |
| "loss": 0.1374, |
| "step": 5600 |
| }, |
| { |
| "epoch": 26.51, |
| "learning_rate": 3.523884103367267e-05, |
| "loss": 0.133, |
| "step": 5700 |
| }, |
| { |
| "epoch": 26.98, |
| "learning_rate": 3.0540328895849644e-05, |
| "loss": 0.1325, |
| "step": 5800 |
| }, |
| { |
| "epoch": 27.44, |
| "learning_rate": 2.584181675802662e-05, |
| "loss": 0.1324, |
| "step": 5900 |
| }, |
| { |
| "epoch": 27.91, |
| "learning_rate": 2.1143304620203598e-05, |
| "loss": 0.1255, |
| "step": 6000 |
| }, |
| { |
| "epoch": 27.91, |
| "eval_cer": 0.08848641060456125, |
| "eval_loss": 0.49347782135009766, |
| "eval_runtime": 23.1133, |
| "eval_samples_per_second": 75.195, |
| "eval_steps_per_second": 4.716, |
| "eval_wer": 0.19209351050339085, |
| "step": 6000 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 6450, |
| "num_train_epochs": 30, |
| "save_steps": 600, |
| "total_flos": 4.161559358313857e+19, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
