{ "best_metric": 0.9694027655192704, "best_model_checkpoint": "wav2vec2-base-ft-keyword-spotting/checkpoint-995", "epoch": 4.981226533166458, "eval_steps": 500, "global_step": 995, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.05, "learning_rate": 3e-06, "loss": 2.4885, "step": 10 }, { "epoch": 0.1, "learning_rate": 6e-06, "loss": 2.461, "step": 20 }, { "epoch": 0.15, "learning_rate": 9e-06, "loss": 2.404, "step": 30 }, { "epoch": 0.2, "learning_rate": 1.2e-05, "loss": 2.2766, "step": 40 }, { "epoch": 0.25, "learning_rate": 1.5e-05, "loss": 2.0127, "step": 50 }, { "epoch": 0.3, "learning_rate": 1.8e-05, "loss": 1.8231, "step": 60 }, { "epoch": 0.35, "learning_rate": 2.1e-05, "loss": 1.7101, "step": 70 }, { "epoch": 0.4, "learning_rate": 2.4e-05, "loss": 1.6402, "step": 80 }, { "epoch": 0.45, "learning_rate": 2.7000000000000002e-05, "loss": 1.5429, "step": 90 }, { "epoch": 0.5, "learning_rate": 3e-05, "loss": 1.4995, "step": 100 }, { "epoch": 0.55, "learning_rate": 2.9664804469273744e-05, "loss": 1.522, "step": 110 }, { "epoch": 0.6, "learning_rate": 2.9329608938547488e-05, "loss": 1.5216, "step": 120 }, { "epoch": 0.65, "learning_rate": 2.899441340782123e-05, "loss": 1.496, "step": 130 }, { "epoch": 0.7, "learning_rate": 2.8659217877094975e-05, "loss": 1.4857, "step": 140 }, { "epoch": 0.75, "learning_rate": 2.8324022346368715e-05, "loss": 1.4528, "step": 150 }, { "epoch": 0.8, "learning_rate": 2.798882681564246e-05, "loss": 1.4026, "step": 160 }, { "epoch": 0.85, "learning_rate": 2.7653631284916202e-05, "loss": 1.4524, "step": 170 }, { "epoch": 0.9, "learning_rate": 2.7318435754189945e-05, "loss": 1.3755, "step": 180 }, { "epoch": 0.95, "learning_rate": 2.698324022346369e-05, "loss": 1.3203, "step": 190 }, { "epoch": 1.0, "eval_accuracy": 0.6328331862312445, "eval_loss": 1.2906286716461182, "eval_runtime": 7.8945, "eval_samples_per_second": 861.104, "eval_steps_per_second": 13.554, "step": 199 }, { "epoch": 1.0, "learning_rate": 2.6648044692737432e-05, "loss": 1.3231, "step": 200 }, { "epoch": 1.05, "learning_rate": 2.6312849162011176e-05, "loss": 1.2471, "step": 210 }, { "epoch": 1.1, "learning_rate": 2.597765363128492e-05, "loss": 1.2807, "step": 220 }, { "epoch": 1.15, "learning_rate": 2.564245810055866e-05, "loss": 1.2235, "step": 230 }, { "epoch": 1.2, "learning_rate": 2.5307262569832403e-05, "loss": 1.2313, "step": 240 }, { "epoch": 1.25, "learning_rate": 2.4972067039106143e-05, "loss": 1.1449, "step": 250 }, { "epoch": 1.3, "learning_rate": 2.4636871508379887e-05, "loss": 1.1987, "step": 260 }, { "epoch": 1.35, "learning_rate": 2.430167597765363e-05, "loss": 1.1639, "step": 270 }, { "epoch": 1.4, "learning_rate": 2.3966480446927374e-05, "loss": 1.0966, "step": 280 }, { "epoch": 1.45, "learning_rate": 2.3631284916201117e-05, "loss": 1.1318, "step": 290 }, { "epoch": 1.5, "learning_rate": 2.329608938547486e-05, "loss": 1.0765, "step": 300 }, { "epoch": 1.55, "learning_rate": 2.2960893854748604e-05, "loss": 1.0647, "step": 310 }, { "epoch": 1.6, "learning_rate": 2.2625698324022348e-05, "loss": 0.9996, "step": 320 }, { "epoch": 1.65, "learning_rate": 2.229050279329609e-05, "loss": 1.0009, "step": 330 }, { "epoch": 1.7, "learning_rate": 2.195530726256983e-05, "loss": 1.0277, "step": 340 }, { "epoch": 1.75, "learning_rate": 2.1620111731843575e-05, "loss": 1.0153, "step": 350 }, { "epoch": 1.8, "learning_rate": 2.1284916201117318e-05, "loss": 1.0128, "step": 360 }, { "epoch": 1.85, "learning_rate": 2.0949720670391062e-05, "loss": 0.9423, "step": 370 }, { "epoch": 1.9, "learning_rate": 2.0614525139664805e-05, "loss": 0.9471, "step": 380 }, { "epoch": 1.95, "learning_rate": 2.027932960893855e-05, "loss": 0.9587, "step": 390 }, { "epoch": 2.0, "eval_accuracy": 0.7355104442483084, "eval_loss": 0.7793017029762268, "eval_runtime": 7.9323, "eval_samples_per_second": 857.006, "eval_steps_per_second": 13.489, "step": 399 }, { "epoch": 2.0, "learning_rate": 1.9944134078212292e-05, "loss": 0.9263, "step": 400 }, { "epoch": 2.05, "learning_rate": 1.9608938547486036e-05, "loss": 0.8469, "step": 410 }, { "epoch": 2.1, "learning_rate": 1.927374301675978e-05, "loss": 0.8733, "step": 420 }, { "epoch": 2.15, "learning_rate": 1.893854748603352e-05, "loss": 0.85, "step": 430 }, { "epoch": 2.2, "learning_rate": 1.8603351955307263e-05, "loss": 0.84, "step": 440 }, { "epoch": 2.25, "learning_rate": 1.8268156424581006e-05, "loss": 0.8049, "step": 450 }, { "epoch": 2.3, "learning_rate": 1.793296089385475e-05, "loss": 0.7884, "step": 460 }, { "epoch": 2.35, "learning_rate": 1.7597765363128493e-05, "loss": 0.8033, "step": 470 }, { "epoch": 2.4, "learning_rate": 1.7262569832402237e-05, "loss": 0.7183, "step": 480 }, { "epoch": 2.45, "learning_rate": 1.6927374301675977e-05, "loss": 0.7245, "step": 490 }, { "epoch": 2.5, "learning_rate": 1.659217877094972e-05, "loss": 0.6581, "step": 500 }, { "epoch": 2.55, "learning_rate": 1.6256983240223464e-05, "loss": 0.6653, "step": 510 }, { "epoch": 2.6, "learning_rate": 1.5921787709497207e-05, "loss": 0.6699, "step": 520 }, { "epoch": 2.65, "learning_rate": 1.5586592178770948e-05, "loss": 0.6635, "step": 530 }, { "epoch": 2.7, "learning_rate": 1.5251396648044693e-05, "loss": 0.6372, "step": 540 }, { "epoch": 2.75, "learning_rate": 1.4916201117318435e-05, "loss": 0.6368, "step": 550 }, { "epoch": 2.8, "learning_rate": 1.4581005586592178e-05, "loss": 0.5918, "step": 560 }, { "epoch": 2.85, "learning_rate": 1.4245810055865922e-05, "loss": 0.6305, "step": 570 }, { "epoch": 2.9, "learning_rate": 1.3910614525139665e-05, "loss": 0.5916, "step": 580 }, { "epoch": 2.95, "learning_rate": 1.3575418994413409e-05, "loss": 0.6218, "step": 590 }, { "epoch": 3.0, "eval_accuracy": 0.9289496910856134, "eval_loss": 0.38581380248069763, "eval_runtime": 7.9106, "eval_samples_per_second": 859.357, "eval_steps_per_second": 13.526, "step": 599 }, { "epoch": 3.0, "learning_rate": 1.324022346368715e-05, "loss": 0.5873, "step": 600 }, { "epoch": 3.05, "learning_rate": 1.2905027932960894e-05, "loss": 0.5561, "step": 610 }, { "epoch": 3.1, "learning_rate": 1.2569832402234637e-05, "loss": 0.5657, "step": 620 }, { "epoch": 3.15, "learning_rate": 1.2234636871508381e-05, "loss": 0.5355, "step": 630 }, { "epoch": 3.2, "learning_rate": 1.1899441340782124e-05, "loss": 0.5338, "step": 640 }, { "epoch": 3.25, "learning_rate": 1.1564245810055866e-05, "loss": 0.5436, "step": 650 }, { "epoch": 3.3, "learning_rate": 1.122905027932961e-05, "loss": 0.5289, "step": 660 }, { "epoch": 3.35, "learning_rate": 1.0893854748603353e-05, "loss": 0.4941, "step": 670 }, { "epoch": 3.4, "learning_rate": 1.0558659217877095e-05, "loss": 0.46, "step": 680 }, { "epoch": 3.45, "learning_rate": 1.0223463687150837e-05, "loss": 0.4775, "step": 690 }, { "epoch": 3.5, "learning_rate": 9.88826815642458e-06, "loss": 0.4725, "step": 700 }, { "epoch": 3.55, "learning_rate": 9.553072625698324e-06, "loss": 0.5098, "step": 710 }, { "epoch": 3.6, "learning_rate": 9.217877094972067e-06, "loss": 0.4217, "step": 720 }, { "epoch": 3.65, "learning_rate": 8.88268156424581e-06, "loss": 0.4628, "step": 730 }, { "epoch": 3.7, "learning_rate": 8.547486033519553e-06, "loss": 0.4573, "step": 740 }, { "epoch": 3.75, "learning_rate": 8.212290502793296e-06, "loss": 0.4078, "step": 750 }, { "epoch": 3.8, "learning_rate": 7.87709497206704e-06, "loss": 0.4339, "step": 760 }, { "epoch": 3.85, "learning_rate": 7.541899441340783e-06, "loss": 0.4426, "step": 770 }, { "epoch": 3.9, "learning_rate": 7.206703910614525e-06, "loss": 0.4318, "step": 780 }, { "epoch": 3.95, "learning_rate": 6.871508379888268e-06, "loss": 0.4379, "step": 790 }, { "epoch": 4.0, "eval_accuracy": 0.9688143571638718, "eval_loss": 0.2581191658973694, "eval_runtime": 7.8989, "eval_samples_per_second": 860.625, "eval_steps_per_second": 13.546, "step": 799 }, { "epoch": 4.01, "learning_rate": 6.536312849162011e-06, "loss": 0.4175, "step": 800 }, { "epoch": 4.06, "learning_rate": 6.2011173184357546e-06, "loss": 0.4289, "step": 810 }, { "epoch": 4.11, "learning_rate": 5.865921787709497e-06, "loss": 0.4106, "step": 820 }, { "epoch": 4.16, "learning_rate": 5.530726256983241e-06, "loss": 0.4042, "step": 830 }, { "epoch": 4.21, "learning_rate": 5.195530726256983e-06, "loss": 0.3968, "step": 840 }, { "epoch": 4.26, "learning_rate": 4.860335195530726e-06, "loss": 0.4241, "step": 850 }, { "epoch": 4.31, "learning_rate": 4.5251396648044695e-06, "loss": 0.4201, "step": 860 }, { "epoch": 4.36, "learning_rate": 4.189944134078212e-06, "loss": 0.4136, "step": 870 }, { "epoch": 4.41, "learning_rate": 3.854748603351956e-06, "loss": 0.4093, "step": 880 }, { "epoch": 4.46, "learning_rate": 3.5195530726256983e-06, "loss": 0.4178, "step": 890 }, { "epoch": 4.51, "learning_rate": 3.1843575418994414e-06, "loss": 0.3721, "step": 900 }, { "epoch": 4.56, "learning_rate": 2.8491620111731845e-06, "loss": 0.3936, "step": 910 }, { "epoch": 4.61, "learning_rate": 2.513966480446927e-06, "loss": 0.3675, "step": 920 }, { "epoch": 4.66, "learning_rate": 2.1787709497206706e-06, "loss": 0.4145, "step": 930 }, { "epoch": 4.71, "learning_rate": 1.8435754189944133e-06, "loss": 0.4089, "step": 940 }, { "epoch": 4.76, "learning_rate": 1.5083798882681566e-06, "loss": 0.3767, "step": 950 }, { "epoch": 4.81, "learning_rate": 1.1731843575418994e-06, "loss": 0.3827, "step": 960 }, { "epoch": 4.86, "learning_rate": 8.379888268156424e-07, "loss": 0.3819, "step": 970 }, { "epoch": 4.91, "learning_rate": 5.027932960893855e-07, "loss": 0.3897, "step": 980 }, { "epoch": 4.96, "learning_rate": 1.675977653631285e-07, "loss": 0.3779, "step": 990 }, { "epoch": 4.98, "eval_accuracy": 0.9694027655192704, "eval_loss": 0.22695787250995636, "eval_runtime": 7.9078, "eval_samples_per_second": 859.662, "eval_steps_per_second": 13.531, "step": 995 }, { "epoch": 4.98, "step": 995, "total_flos": 2.311202904603648e+18, "train_loss": 0.878907411182346, "train_runtime": 597.8339, "train_samples_per_second": 427.326, "train_steps_per_second": 1.664 } ], "logging_steps": 10, "max_steps": 995, "num_train_epochs": 5, "save_steps": 500, "total_flos": 2.311202904603648e+18, "trial_name": null, "trial_params": null }