| { | |
| "best_metric": 0.9442970822281167, | |
| "best_model_checkpoint": "wav2vec2-large/checkpoint-11500", | |
| "epoch": 31.998484082870135, | |
| "global_step": 15808, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.487666034155598e-06, | |
| "loss": 4.6314, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_accuracy": 0.02046229632436529, | |
| "eval_f1": 0.004873418863566646, | |
| "eval_loss": 4.916522026062012, | |
| "eval_precision": 0.0027763277100147127, | |
| "eval_runtime": 123.5513, | |
| "eval_samples_per_second": 21.36, | |
| "eval_steps_per_second": 1.781, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 1.8975332068311197e-05, | |
| "loss": 3.7739, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_accuracy": 0.03561955286093217, | |
| "eval_f1": 0.02524670069279936, | |
| "eval_loss": 4.449059009552002, | |
| "eval_precision": 0.0750288473053048, | |
| "eval_runtime": 121.7021, | |
| "eval_samples_per_second": 21.684, | |
| "eval_steps_per_second": 1.808, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 2.846299810246679e-05, | |
| "loss": 2.5035, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "eval_accuracy": 0.11292156119742326, | |
| "eval_f1": 0.11144842685774854, | |
| "eval_loss": 4.142854690551758, | |
| "eval_precision": 0.2671650922898445, | |
| "eval_runtime": 120.9962, | |
| "eval_samples_per_second": 21.811, | |
| "eval_steps_per_second": 1.818, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 2.9116468686300697e-05, | |
| "loss": 1.5633, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "eval_accuracy": 0.36756347101174686, | |
| "eval_f1": 0.3830364851222239, | |
| "eval_loss": 3.1973092555999756, | |
| "eval_precision": 0.6598356853987439, | |
| "eval_runtime": 122.0663, | |
| "eval_samples_per_second": 21.619, | |
| "eval_steps_per_second": 1.802, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 2.8062135376396992e-05, | |
| "loss": 1.0538, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "eval_accuracy": 0.5888594164456233, | |
| "eval_f1": 0.6556643570555963, | |
| "eval_loss": 2.5479371547698975, | |
| "eval_precision": 0.8416636658887603, | |
| "eval_runtime": 121.3195, | |
| "eval_samples_per_second": 21.752, | |
| "eval_steps_per_second": 1.813, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 2.700780206649329e-05, | |
| "loss": 0.7422, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "eval_accuracy": 0.7824933687002652, | |
| "eval_f1": 0.8194352733059617, | |
| "eval_loss": 1.4494293928146362, | |
| "eval_precision": 0.8920845853165854, | |
| "eval_runtime": 120.4455, | |
| "eval_samples_per_second": 21.91, | |
| "eval_steps_per_second": 1.827, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 2.5953468756589585e-05, | |
| "loss": 0.5762, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "eval_accuracy": 0.7726411519514967, | |
| "eval_f1": 0.8267111468977018, | |
| "eval_loss": 1.316819190979004, | |
| "eval_precision": 0.9277444722087854, | |
| "eval_runtime": 122.3892, | |
| "eval_samples_per_second": 21.562, | |
| "eval_steps_per_second": 1.798, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 2.489913544668588e-05, | |
| "loss": 0.46, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "eval_accuracy": 0.8563849943160288, | |
| "eval_f1": 0.898213371629901, | |
| "eval_loss": 0.8782555460929871, | |
| "eval_precision": 0.9531998814815654, | |
| "eval_runtime": 120.9059, | |
| "eval_samples_per_second": 21.827, | |
| "eval_steps_per_second": 1.82, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "learning_rate": 2.3844802136782175e-05, | |
| "loss": 0.4007, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "eval_accuracy": 0.8738158393330807, | |
| "eval_f1": 0.913683470937564, | |
| "eval_loss": 0.752358615398407, | |
| "eval_precision": 0.9636991964706034, | |
| "eval_runtime": 120.9904, | |
| "eval_samples_per_second": 21.812, | |
| "eval_steps_per_second": 1.818, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "learning_rate": 2.279046882687847e-05, | |
| "loss": 0.3374, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "eval_accuracy": 0.8851837817355058, | |
| "eval_f1": 0.9220598204792498, | |
| "eval_loss": 0.6385864615440369, | |
| "eval_precision": 0.9677694535212527, | |
| "eval_runtime": 120.467, | |
| "eval_samples_per_second": 21.906, | |
| "eval_steps_per_second": 1.826, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "learning_rate": 2.1736135516974768e-05, | |
| "loss": 0.3108, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "eval_accuracy": 0.9105721864342554, | |
| "eval_f1": 0.9372909559856852, | |
| "eval_loss": 0.5049420595169067, | |
| "eval_precision": 0.9680582892579286, | |
| "eval_runtime": 121.8653, | |
| "eval_samples_per_second": 21.655, | |
| "eval_steps_per_second": 1.805, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "learning_rate": 2.0681802207071063e-05, | |
| "loss": 0.2735, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "eval_accuracy": 0.8904888215233043, | |
| "eval_f1": 0.9225848099506897, | |
| "eval_loss": 0.6096966862678528, | |
| "eval_precision": 0.9623634379449385, | |
| "eval_runtime": 121.4949, | |
| "eval_samples_per_second": 21.721, | |
| "eval_steps_per_second": 1.811, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "learning_rate": 1.9627468897167357e-05, | |
| "loss": 0.2716, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "eval_accuracy": 0.8999621068586586, | |
| "eval_f1": 0.9206365230401724, | |
| "eval_loss": 0.4542546570301056, | |
| "eval_precision": 0.9569343688614892, | |
| "eval_runtime": 120.692, | |
| "eval_samples_per_second": 21.866, | |
| "eval_steps_per_second": 1.823, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 14.17, | |
| "learning_rate": 1.8573135587263652e-05, | |
| "loss": 0.2484, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 14.17, | |
| "eval_accuracy": 0.9272451686244789, | |
| "eval_f1": 0.9489393378254958, | |
| "eval_loss": 0.3964819610118866, | |
| "eval_precision": 0.9742030978909282, | |
| "eval_runtime": 121.4363, | |
| "eval_samples_per_second": 21.732, | |
| "eval_steps_per_second": 1.812, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 15.18, | |
| "learning_rate": 1.751880227735995e-05, | |
| "loss": 0.228, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 15.18, | |
| "eval_accuracy": 0.8855627131489201, | |
| "eval_f1": 0.9256828810141209, | |
| "eval_loss": 0.6806507110595703, | |
| "eval_precision": 0.9777330671722411, | |
| "eval_runtime": 122.0415, | |
| "eval_samples_per_second": 21.624, | |
| "eval_steps_per_second": 1.803, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 16.19, | |
| "learning_rate": 1.6464468967456245e-05, | |
| "loss": 0.2307, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 16.19, | |
| "eval_accuracy": 0.9173929518757105, | |
| "eval_f1": 0.946436431649072, | |
| "eval_loss": 0.5219454765319824, | |
| "eval_precision": 0.9801938448683835, | |
| "eval_runtime": 120.3841, | |
| "eval_samples_per_second": 21.921, | |
| "eval_steps_per_second": 1.827, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 17.21, | |
| "learning_rate": 1.541013565755254e-05, | |
| "loss": 0.2169, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 17.21, | |
| "eval_accuracy": 0.9120879120879121, | |
| "eval_f1": 0.9337616922587397, | |
| "eval_loss": 0.4630146920681, | |
| "eval_precision": 0.9677088927055834, | |
| "eval_runtime": 121.7745, | |
| "eval_samples_per_second": 21.671, | |
| "eval_steps_per_second": 1.807, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 18.22, | |
| "learning_rate": 1.4355802347648837e-05, | |
| "loss": 0.1997, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 18.22, | |
| "eval_accuracy": 0.9128457749147404, | |
| "eval_f1": 0.9397548887949031, | |
| "eval_loss": 0.5151729583740234, | |
| "eval_precision": 0.9740319117972248, | |
| "eval_runtime": 121.0656, | |
| "eval_samples_per_second": 21.798, | |
| "eval_steps_per_second": 1.817, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "learning_rate": 1.3301469037745133e-05, | |
| "loss": 0.1921, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "eval_accuracy": 0.9143615005683972, | |
| "eval_f1": 0.9475938884984938, | |
| "eval_loss": 0.510482668876648, | |
| "eval_precision": 0.9867250624910348, | |
| "eval_runtime": 122.1644, | |
| "eval_samples_per_second": 21.602, | |
| "eval_steps_per_second": 1.801, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 20.24, | |
| "learning_rate": 1.2247135727841428e-05, | |
| "loss": 0.1825, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 20.24, | |
| "eval_accuracy": 0.905267146646457, | |
| "eval_f1": 0.9406773355537659, | |
| "eval_loss": 0.6302403807640076, | |
| "eval_precision": 0.9831536512710705, | |
| "eval_runtime": 121.7536, | |
| "eval_samples_per_second": 21.675, | |
| "eval_steps_per_second": 1.807, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "learning_rate": 1.1192802417937724e-05, | |
| "loss": 0.1786, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "eval_accuracy": 0.9272451686244789, | |
| "eval_f1": 0.952370485075943, | |
| "eval_loss": 0.460215300321579, | |
| "eval_precision": 0.9813011833529409, | |
| "eval_runtime": 124.2398, | |
| "eval_samples_per_second": 21.241, | |
| "eval_steps_per_second": 1.771, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 22.27, | |
| "learning_rate": 1.013846910803402e-05, | |
| "loss": 0.1671, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 22.27, | |
| "eval_accuracy": 0.9147404319818113, | |
| "eval_f1": 0.9444178129216141, | |
| "eval_loss": 0.544323205947876, | |
| "eval_precision": 0.9794231668433278, | |
| "eval_runtime": 120.9552, | |
| "eval_samples_per_second": 21.818, | |
| "eval_steps_per_second": 1.819, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 23.28, | |
| "learning_rate": 9.084135798130316e-06, | |
| "loss": 0.1623, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 23.28, | |
| "eval_accuracy": 0.9442970822281167, | |
| "eval_f1": 0.9604353051727885, | |
| "eval_loss": 0.3412924110889435, | |
| "eval_precision": 0.9780145058039889, | |
| "eval_runtime": 122.6656, | |
| "eval_samples_per_second": 21.514, | |
| "eval_steps_per_second": 1.793, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 24.29, | |
| "learning_rate": 8.029802488226612e-06, | |
| "loss": 0.1595, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 24.29, | |
| "eval_accuracy": 0.9287608942781357, | |
| "eval_f1": 0.9531334781334764, | |
| "eval_loss": 0.44775113463401794, | |
| "eval_precision": 0.9812894876888261, | |
| "eval_runtime": 120.6582, | |
| "eval_samples_per_second": 21.872, | |
| "eval_steps_per_second": 1.823, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 25.3, | |
| "learning_rate": 6.975469178322908e-06, | |
| "loss": 0.151, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 25.3, | |
| "eval_accuracy": 0.9359605911330049, | |
| "eval_f1": 0.9571224982425287, | |
| "eval_loss": 0.41782599687576294, | |
| "eval_precision": 0.9818213438674759, | |
| "eval_runtime": 121.5443, | |
| "eval_samples_per_second": 21.712, | |
| "eval_steps_per_second": 1.81, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "learning_rate": 5.9211358684192026e-06, | |
| "loss": 0.1472, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "eval_accuracy": 0.9355816597195907, | |
| "eval_f1": 0.9577691906088884, | |
| "eval_loss": 0.41541919112205505, | |
| "eval_precision": 0.9832631333451843, | |
| "eval_runtime": 122.2041, | |
| "eval_samples_per_second": 21.595, | |
| "eval_steps_per_second": 1.8, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 27.33, | |
| "learning_rate": 4.866802558515498e-06, | |
| "loss": 0.1473, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 27.33, | |
| "eval_accuracy": 0.931792345585449, | |
| "eval_f1": 0.9560631584126075, | |
| "eval_loss": 0.4548525810241699, | |
| "eval_precision": 0.9837464944410116, | |
| "eval_runtime": 121.1061, | |
| "eval_samples_per_second": 21.791, | |
| "eval_steps_per_second": 1.817, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 28.34, | |
| "learning_rate": 3.8124692486117947e-06, | |
| "loss": 0.131, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 28.34, | |
| "eval_accuracy": 0.9424024251610459, | |
| "eval_f1": 0.9620998995027674, | |
| "eval_loss": 0.3574429154396057, | |
| "eval_precision": 0.984478231442749, | |
| "eval_runtime": 120.0382, | |
| "eval_samples_per_second": 21.985, | |
| "eval_steps_per_second": 1.833, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 29.35, | |
| "learning_rate": 2.7581359387080904e-06, | |
| "loss": 0.134, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 29.35, | |
| "eval_accuracy": 0.9333080712391058, | |
| "eval_f1": 0.9568214460159893, | |
| "eval_loss": 0.4474771022796631, | |
| "eval_precision": 0.9840495545740223, | |
| "eval_runtime": 123.1013, | |
| "eval_samples_per_second": 21.438, | |
| "eval_steps_per_second": 1.787, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 30.36, | |
| "learning_rate": 1.7038026288043862e-06, | |
| "loss": 0.1282, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 30.36, | |
| "eval_accuracy": 0.93823417961349, | |
| "eval_f1": 0.959091295015073, | |
| "eval_loss": 0.401244193315506, | |
| "eval_precision": 0.9836786405574566, | |
| "eval_runtime": 121.7667, | |
| "eval_samples_per_second": 21.673, | |
| "eval_steps_per_second": 1.807, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 31.38, | |
| "learning_rate": 6.494693189006819e-07, | |
| "loss": 0.1307, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 31.38, | |
| "eval_accuracy": 0.94278135657446, | |
| "eval_f1": 0.9623515495402261, | |
| "eval_loss": 0.35516688227653503, | |
| "eval_precision": 0.9847235247308098, | |
| "eval_runtime": 121.4163, | |
| "eval_samples_per_second": 21.735, | |
| "eval_steps_per_second": 1.812, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "step": 15808, | |
| "total_flos": 1.2696466131100762e+20, | |
| "train_loss": 0.6361874522950485, | |
| "train_runtime": 48273.5986, | |
| "train_samples_per_second": 15.742, | |
| "train_steps_per_second": 0.327 | |
| } | |
| ], | |
| "max_steps": 15808, | |
| "num_train_epochs": 32, | |
| "total_flos": 1.2696466131100762e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |