| { | |
| "best_metric": 0.41899441340782123, | |
| "best_model_checkpoint": "convnextv2-base-1k-224-for-pre_evaluation/checkpoint-128", | |
| "epoch": 30.0, | |
| "eval_steps": 500, | |
| "global_step": 480, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 1.6, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.29608938547486036, | |
| "eval_loss": 1.5315604209899902, | |
| "eval_runtime": 6.8342, | |
| "eval_samples_per_second": 52.384, | |
| "eval_steps_per_second": 1.756, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 1.535, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 3.125e-05, | |
| "loss": 1.5084, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.2849162011173184, | |
| "eval_loss": 1.5060781240463257, | |
| "eval_runtime": 7.4696, | |
| "eval_samples_per_second": 47.928, | |
| "eval_steps_per_second": 1.607, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.5134, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.3240223463687151, | |
| "eval_loss": 1.4968407154083252, | |
| "eval_runtime": 7.1172, | |
| "eval_samples_per_second": 50.301, | |
| "eval_steps_per_second": 1.686, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 4.976851851851852e-05, | |
| "loss": 1.4694, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 1.4663, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.33519553072625696, | |
| "eval_loss": 1.4607384204864502, | |
| "eval_runtime": 7.4013, | |
| "eval_samples_per_second": 48.37, | |
| "eval_steps_per_second": 1.621, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 4.745370370370371e-05, | |
| "loss": 1.4375, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 1.4046, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.3268156424581006, | |
| "eval_loss": 1.4509011507034302, | |
| "eval_runtime": 6.83, | |
| "eval_samples_per_second": 52.416, | |
| "eval_steps_per_second": 1.757, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "learning_rate": 4.5138888888888894e-05, | |
| "loss": 1.4085, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.388268156424581, | |
| "eval_loss": 1.4423185586929321, | |
| "eval_runtime": 7.3683, | |
| "eval_samples_per_second": 48.587, | |
| "eval_steps_per_second": 1.629, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 4.3981481481481486e-05, | |
| "loss": 1.3765, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 4.282407407407408e-05, | |
| "loss": 1.3443, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.4022346368715084, | |
| "eval_loss": 1.400512456893921, | |
| "eval_runtime": 6.9156, | |
| "eval_samples_per_second": 51.767, | |
| "eval_steps_per_second": 1.735, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.3025, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.41899441340782123, | |
| "eval_loss": 1.359870195388794, | |
| "eval_runtime": 7.041, | |
| "eval_samples_per_second": 50.845, | |
| "eval_steps_per_second": 1.704, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 4.0509259259259265e-05, | |
| "loss": 1.2668, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "learning_rate": 3.935185185185186e-05, | |
| "loss": 1.2627, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.39106145251396646, | |
| "eval_loss": 1.3638169765472412, | |
| "eval_runtime": 7.2532, | |
| "eval_samples_per_second": 49.357, | |
| "eval_steps_per_second": 1.654, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 3.8194444444444444e-05, | |
| "loss": 1.2427, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 1.2099, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.34916201117318435, | |
| "eval_loss": 1.4057648181915283, | |
| "eval_runtime": 7.508, | |
| "eval_samples_per_second": 47.682, | |
| "eval_steps_per_second": 1.598, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 3.587962962962963e-05, | |
| "loss": 1.2086, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.3407821229050279, | |
| "eval_loss": 1.443146824836731, | |
| "eval_runtime": 7.4283, | |
| "eval_samples_per_second": 48.194, | |
| "eval_steps_per_second": 1.615, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 1.1684, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 11.88, | |
| "learning_rate": 3.3564814814814815e-05, | |
| "loss": 1.1393, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.34916201117318435, | |
| "eval_loss": 1.4143450260162354, | |
| "eval_runtime": 7.1396, | |
| "eval_samples_per_second": 50.143, | |
| "eval_steps_per_second": 1.681, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 1.1039, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.388268156424581, | |
| "eval_loss": 1.4305065870285034, | |
| "eval_runtime": 6.8442, | |
| "eval_samples_per_second": 52.307, | |
| "eval_steps_per_second": 1.753, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 3.125e-05, | |
| "loss": 1.0641, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "learning_rate": 3.0092592592592593e-05, | |
| "loss": 1.0551, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.35195530726256985, | |
| "eval_loss": 1.5202596187591553, | |
| "eval_runtime": 7.2925, | |
| "eval_samples_per_second": 49.091, | |
| "eval_steps_per_second": 1.646, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 14.38, | |
| "learning_rate": 2.8935185185185186e-05, | |
| "loss": 1.0686, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 1.0368, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.3324022346368715, | |
| "eval_loss": 1.5117393732070923, | |
| "eval_runtime": 6.8377, | |
| "eval_samples_per_second": 52.356, | |
| "eval_steps_per_second": 1.755, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 15.62, | |
| "learning_rate": 2.6620370370370372e-05, | |
| "loss": 0.9753, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.3770949720670391, | |
| "eval_loss": 1.4545259475708008, | |
| "eval_runtime": 7.0796, | |
| "eval_samples_per_second": 50.568, | |
| "eval_steps_per_second": 1.695, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "learning_rate": 2.5462962962962965e-05, | |
| "loss": 0.9677, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 16.88, | |
| "learning_rate": 2.4305555555555558e-05, | |
| "loss": 0.938, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.33519553072625696, | |
| "eval_loss": 1.5396308898925781, | |
| "eval_runtime": 7.1168, | |
| "eval_samples_per_second": 50.304, | |
| "eval_steps_per_second": 1.686, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.899, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.3407821229050279, | |
| "eval_loss": 1.5770219564437866, | |
| "eval_runtime": 6.8822, | |
| "eval_samples_per_second": 52.018, | |
| "eval_steps_per_second": 1.744, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "learning_rate": 2.1990740740740743e-05, | |
| "loss": 0.9047, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.8629, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.3128491620111732, | |
| "eval_loss": 1.7105906009674072, | |
| "eval_runtime": 7.3502, | |
| "eval_samples_per_second": 48.706, | |
| "eval_steps_per_second": 1.633, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 19.38, | |
| "learning_rate": 1.967592592592593e-05, | |
| "loss": 0.8624, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.8674, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.33519553072625696, | |
| "eval_loss": 1.5864217281341553, | |
| "eval_runtime": 7.1963, | |
| "eval_samples_per_second": 49.748, | |
| "eval_steps_per_second": 1.668, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 20.62, | |
| "learning_rate": 1.736111111111111e-05, | |
| "loss": 0.7789, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.3407821229050279, | |
| "eval_loss": 1.6129050254821777, | |
| "eval_runtime": 7.1496, | |
| "eval_samples_per_second": 50.073, | |
| "eval_steps_per_second": 1.678, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "learning_rate": 1.6203703703703704e-05, | |
| "loss": 0.8161, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 21.88, | |
| "learning_rate": 1.5046296296296297e-05, | |
| "loss": 0.7426, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.36033519553072624, | |
| "eval_loss": 1.6353477239608765, | |
| "eval_runtime": 7.4456, | |
| "eval_samples_per_second": 48.082, | |
| "eval_steps_per_second": 1.612, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 22.5, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.7677, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.3463687150837989, | |
| "eval_loss": 1.6793445348739624, | |
| "eval_runtime": 6.994, | |
| "eval_samples_per_second": 51.187, | |
| "eval_steps_per_second": 1.716, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 23.12, | |
| "learning_rate": 1.2731481481481482e-05, | |
| "loss": 0.7327, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 23.75, | |
| "learning_rate": 1.1574074074074075e-05, | |
| "loss": 0.7172, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.3575418994413408, | |
| "eval_loss": 1.6759321689605713, | |
| "eval_runtime": 7.4394, | |
| "eval_samples_per_second": 48.122, | |
| "eval_steps_per_second": 1.613, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 24.38, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 0.6759, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.6809, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.3659217877094972, | |
| "eval_loss": 1.701292634010315, | |
| "eval_runtime": 7.4138, | |
| "eval_samples_per_second": 48.288, | |
| "eval_steps_per_second": 1.619, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 25.62, | |
| "learning_rate": 8.101851851851852e-06, | |
| "loss": 0.6619, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.36312849162011174, | |
| "eval_loss": 1.7108293771743774, | |
| "eval_runtime": 7.238, | |
| "eval_samples_per_second": 49.461, | |
| "eval_steps_per_second": 1.658, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 26.25, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.6773, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 26.88, | |
| "learning_rate": 5.787037037037038e-06, | |
| "loss": 0.6656, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.3715083798882682, | |
| "eval_loss": 1.7327028512954712, | |
| "eval_runtime": 6.8416, | |
| "eval_samples_per_second": 52.327, | |
| "eval_steps_per_second": 1.754, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 27.5, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.6258, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.3547486033519553, | |
| "eval_loss": 1.7377949953079224, | |
| "eval_runtime": 7.2785, | |
| "eval_samples_per_second": 49.186, | |
| "eval_steps_per_second": 1.649, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 28.12, | |
| "learning_rate": 3.4722222222222224e-06, | |
| "loss": 0.6646, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 28.75, | |
| "learning_rate": 2.3148148148148148e-06, | |
| "loss": 0.6173, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.36033519553072624, | |
| "eval_loss": 1.7461235523223877, | |
| "eval_runtime": 6.8622, | |
| "eval_samples_per_second": 52.17, | |
| "eval_steps_per_second": 1.749, | |
| "step": 464 | |
| }, | |
| { | |
| "epoch": 29.38, | |
| "learning_rate": 1.1574074074074074e-06, | |
| "loss": 0.6482, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.6214, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.35195530726256985, | |
| "eval_loss": 1.7475444078445435, | |
| "eval_runtime": 7.4355, | |
| "eval_samples_per_second": 48.148, | |
| "eval_steps_per_second": 1.614, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 480, | |
| "total_flos": 4.840276186658304e+18, | |
| "train_loss": 1.0325976332028708, | |
| "train_runtime": 3493.542, | |
| "train_samples_per_second": 17.492, | |
| "train_steps_per_second": 0.137 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 480, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 4.840276186658304e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |