| { | |
| "best_metric": 0.20890280604362488, | |
| "best_model_checkpoint": "wavlm-english-phoneme-v2/checkpoint-7500", | |
| "epoch": 57.69230769230769, | |
| "global_step": 7500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 2.9999999999999997e-05, | |
| "loss": 5.081, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 2.595, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 2.5161, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 2.226, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 0.00015, | |
| "loss": 1.8893, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "eval_cer": 0.7476650184443921, | |
| "eval_loss": 1.3937348127365112, | |
| "eval_runtime": 30.2371, | |
| "eval_samples_per_second": 15.279, | |
| "eval_steps_per_second": 1.918, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 1.1918, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 0.00020999999999999998, | |
| "loss": 0.8122, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.6064, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.92, | |
| "learning_rate": 0.00027, | |
| "loss": 0.4715, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 0.0003, | |
| "loss": 0.3875, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "eval_cer": 0.1790283337257672, | |
| "eval_loss": 0.36919161677360535, | |
| "eval_runtime": 29.9248, | |
| "eval_samples_per_second": 15.439, | |
| "eval_steps_per_second": 1.938, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.46, | |
| "learning_rate": 0.00029558823529411763, | |
| "loss": 0.3297, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.23, | |
| "learning_rate": 0.00029117647058823524, | |
| "loss": 0.2736, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.00028676470588235296, | |
| "loss": 0.2389, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "learning_rate": 0.00028235294117647056, | |
| "loss": 0.1919, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 11.54, | |
| "learning_rate": 0.0002779411764705882, | |
| "loss": 0.1718, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.54, | |
| "eval_cer": 0.11992779216701986, | |
| "eval_loss": 0.2612331211566925, | |
| "eval_runtime": 30.5393, | |
| "eval_samples_per_second": 15.128, | |
| "eval_steps_per_second": 1.899, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.31, | |
| "learning_rate": 0.00027352941176470583, | |
| "loss": 0.1501, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 13.08, | |
| "learning_rate": 0.0002691176470588235, | |
| "loss": 0.1404, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 13.85, | |
| "learning_rate": 0.00026470588235294115, | |
| "loss": 0.1198, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "learning_rate": 0.0002602941176470588, | |
| "loss": 0.1128, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 0.0002558823529411764, | |
| "loss": 0.1022, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "eval_cer": 0.1011694529471784, | |
| "eval_loss": 0.23615849018096924, | |
| "eval_runtime": 37.8684, | |
| "eval_samples_per_second": 12.2, | |
| "eval_steps_per_second": 1.532, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.15, | |
| "learning_rate": 0.0002514705882352941, | |
| "loss": 0.0933, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 16.92, | |
| "learning_rate": 0.00024705882352941174, | |
| "loss": 0.0894, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 17.69, | |
| "learning_rate": 0.0002426470588235294, | |
| "loss": 0.0818, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 18.46, | |
| "learning_rate": 0.000238235294117647, | |
| "loss": 0.0745, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "learning_rate": 0.0002338676470588235, | |
| "loss": 0.0735, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "eval_cer": 0.0909661721999843, | |
| "eval_loss": 0.21879912912845612, | |
| "eval_runtime": 19.7011, | |
| "eval_samples_per_second": 23.451, | |
| "eval_steps_per_second": 2.944, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.00022945588235294116, | |
| "loss": 0.0678, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 20.77, | |
| "learning_rate": 0.0002250441176470588, | |
| "loss": 0.0643, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 21.54, | |
| "learning_rate": 0.00022063235294117646, | |
| "loss": 0.0592, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 22.31, | |
| "learning_rate": 0.0002162205882352941, | |
| "loss": 0.0595, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 23.08, | |
| "learning_rate": 0.00021180882352941175, | |
| "loss": 0.0555, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 23.08, | |
| "eval_cer": 0.08170473275253119, | |
| "eval_loss": 0.22275137901306152, | |
| "eval_runtime": 28.5681, | |
| "eval_samples_per_second": 16.172, | |
| "eval_steps_per_second": 2.03, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 23.85, | |
| "learning_rate": 0.0002073970588235294, | |
| "loss": 0.05, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 24.62, | |
| "learning_rate": 0.00020298529411764705, | |
| "loss": 0.0492, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 25.38, | |
| "learning_rate": 0.00019857352941176468, | |
| "loss": 0.0469, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 26.15, | |
| "learning_rate": 0.00019416176470588234, | |
| "loss": 0.045, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 26.92, | |
| "learning_rate": 0.00018974999999999998, | |
| "loss": 0.0432, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 26.92, | |
| "eval_cer": 0.08044894435287654, | |
| "eval_loss": 0.21576076745986938, | |
| "eval_runtime": 28.5989, | |
| "eval_samples_per_second": 16.154, | |
| "eval_steps_per_second": 2.028, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 27.69, | |
| "learning_rate": 0.00018533823529411764, | |
| "loss": 0.0427, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 28.46, | |
| "learning_rate": 0.00018092647058823527, | |
| "loss": 0.0387, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 29.23, | |
| "learning_rate": 0.00017651470588235293, | |
| "loss": 0.0391, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 0.0001721029411764706, | |
| "loss": 0.0371, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "learning_rate": 0.00016769117647058823, | |
| "loss": 0.0339, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "eval_cer": 0.07707401302880465, | |
| "eval_loss": 0.21580535173416138, | |
| "eval_runtime": 28.7435, | |
| "eval_samples_per_second": 16.073, | |
| "eval_steps_per_second": 2.018, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 31.54, | |
| "learning_rate": 0.0001632794117647059, | |
| "loss": 0.0335, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 32.31, | |
| "learning_rate": 0.00015886764705882352, | |
| "loss": 0.0312, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 33.08, | |
| "learning_rate": 0.00015445588235294118, | |
| "loss": 0.03, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 33.85, | |
| "learning_rate": 0.0001500441176470588, | |
| "loss": 0.0297, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 34.62, | |
| "learning_rate": 0.00014563235294117645, | |
| "loss": 0.0257, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 34.62, | |
| "eval_cer": 0.07487638332940899, | |
| "eval_loss": 0.2141367644071579, | |
| "eval_runtime": 28.3525, | |
| "eval_samples_per_second": 16.295, | |
| "eval_steps_per_second": 2.046, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 35.38, | |
| "learning_rate": 0.0001412205882352941, | |
| "loss": 0.0271, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 36.15, | |
| "learning_rate": 0.00013680882352941175, | |
| "loss": 0.0259, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 36.92, | |
| "learning_rate": 0.0001323970588235294, | |
| "loss": 0.025, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 37.69, | |
| "learning_rate": 0.00012798529411764704, | |
| "loss": 0.0258, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "learning_rate": 0.0001235735294117647, | |
| "loss": 0.0228, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "eval_cer": 0.0705596107055961, | |
| "eval_loss": 0.21314646303653717, | |
| "eval_runtime": 19.2794, | |
| "eval_samples_per_second": 23.963, | |
| "eval_steps_per_second": 3.008, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 39.23, | |
| "learning_rate": 0.00011916176470588235, | |
| "loss": 0.0233, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 0.00011475, | |
| "loss": 0.0204, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 40.77, | |
| "learning_rate": 0.00011033823529411764, | |
| "loss": 0.0194, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 41.54, | |
| "learning_rate": 0.00010592647058823528, | |
| "loss": 0.0167, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 42.31, | |
| "learning_rate": 0.00010151470588235293, | |
| "loss": 0.0189, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 42.31, | |
| "eval_cer": 0.06828349423122204, | |
| "eval_loss": 0.21500647068023682, | |
| "eval_runtime": 19.6812, | |
| "eval_samples_per_second": 23.474, | |
| "eval_steps_per_second": 2.947, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 43.08, | |
| "learning_rate": 9.710294117647057e-05, | |
| "loss": 0.0189, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 43.85, | |
| "learning_rate": 9.269117647058822e-05, | |
| "loss": 0.0182, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 44.62, | |
| "learning_rate": 8.827941176470587e-05, | |
| "loss": 0.0189, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 45.38, | |
| "learning_rate": 8.386764705882352e-05, | |
| "loss": 0.0168, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 46.15, | |
| "learning_rate": 7.945588235294116e-05, | |
| "loss": 0.0141, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 46.15, | |
| "eval_cer": 0.0661643513068048, | |
| "eval_loss": 0.21131280064582825, | |
| "eval_runtime": 20.235, | |
| "eval_samples_per_second": 22.832, | |
| "eval_steps_per_second": 2.866, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 46.92, | |
| "learning_rate": 7.504411764705881e-05, | |
| "loss": 0.0155, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 47.69, | |
| "learning_rate": 7.063235294117646e-05, | |
| "loss": 0.0142, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 48.46, | |
| "learning_rate": 6.62205882352941e-05, | |
| "loss": 0.0139, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 49.23, | |
| "learning_rate": 6.180882352941175e-05, | |
| "loss": 0.0146, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 5.739705882352941e-05, | |
| "loss": 0.0118, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_cer": 0.06388823483243074, | |
| "eval_loss": 0.2127513736486435, | |
| "eval_runtime": 19.4356, | |
| "eval_samples_per_second": 23.771, | |
| "eval_steps_per_second": 2.984, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 50.77, | |
| "learning_rate": 5.298529411764705e-05, | |
| "loss": 0.0117, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 51.54, | |
| "learning_rate": 4.85735294117647e-05, | |
| "loss": 0.0123, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 52.31, | |
| "learning_rate": 4.416176470588235e-05, | |
| "loss": 0.0113, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 53.08, | |
| "learning_rate": 3.975e-05, | |
| "loss": 0.0111, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 53.85, | |
| "learning_rate": 3.5338235294117645e-05, | |
| "loss": 0.011, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 53.85, | |
| "eval_cer": 0.06365277450749549, | |
| "eval_loss": 0.20946960151195526, | |
| "eval_runtime": 19.7474, | |
| "eval_samples_per_second": 23.395, | |
| "eval_steps_per_second": 2.937, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 54.62, | |
| "learning_rate": 3.092647058823529e-05, | |
| "loss": 0.0098, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 55.38, | |
| "learning_rate": 2.651470588235294e-05, | |
| "loss": 0.0105, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 56.15, | |
| "learning_rate": 2.2102941176470584e-05, | |
| "loss": 0.0097, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 56.92, | |
| "learning_rate": 1.7691176470588234e-05, | |
| "loss": 0.0095, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 57.69, | |
| "learning_rate": 1.3279411764705882e-05, | |
| "loss": 0.0097, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 57.69, | |
| "eval_cer": 0.06121968448316459, | |
| "eval_loss": 0.20890280604362488, | |
| "eval_runtime": 20.5541, | |
| "eval_samples_per_second": 22.477, | |
| "eval_steps_per_second": 2.822, | |
| "step": 7500 | |
| } | |
| ], | |
| "max_steps": 7800, | |
| "num_train_epochs": 60, | |
| "total_flos": 2.2797274353269838e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |