| { | |
| "best_metric": 0.981651376146789, | |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-va-U-40/checkpoint-232", | |
| "epoch": 36.12903225806452, | |
| "eval_steps": 500, | |
| "global_step": 280, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.30275229357798167, | |
| "eval_loss": 1.3838695287704468, | |
| "eval_runtime": 1.464, | |
| "eval_samples_per_second": 74.452, | |
| "eval_steps_per_second": 2.732, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 1.4343, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.28440366972477066, | |
| "eval_loss": 1.364293098449707, | |
| "eval_runtime": 0.9466, | |
| "eval_samples_per_second": 115.148, | |
| "eval_steps_per_second": 4.226, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 1.3413, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.3577981651376147, | |
| "eval_loss": 1.3310376405715942, | |
| "eval_runtime": 0.9578, | |
| "eval_samples_per_second": 113.806, | |
| "eval_steps_per_second": 4.176, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 4.960317460317461e-05, | |
| "loss": 1.2473, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5137614678899083, | |
| "eval_loss": 1.1080743074417114, | |
| "eval_runtime": 1.1297, | |
| "eval_samples_per_second": 96.488, | |
| "eval_steps_per_second": 3.541, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_accuracy": 0.7064220183486238, | |
| "eval_loss": 0.8292074203491211, | |
| "eval_runtime": 0.9556, | |
| "eval_samples_per_second": 114.063, | |
| "eval_steps_per_second": 4.186, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 1.0532, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_accuracy": 0.6238532110091743, | |
| "eval_loss": 0.7419529557228088, | |
| "eval_runtime": 0.9843, | |
| "eval_samples_per_second": 110.738, | |
| "eval_steps_per_second": 4.064, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 4.563492063492064e-05, | |
| "loss": 0.917, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.6972477064220184, | |
| "eval_loss": 0.6344928741455078, | |
| "eval_runtime": 0.9762, | |
| "eval_samples_per_second": 111.653, | |
| "eval_steps_per_second": 4.097, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 0.7939, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8532110091743119, | |
| "eval_loss": 0.48984646797180176, | |
| "eval_runtime": 0.9669, | |
| "eval_samples_per_second": 112.734, | |
| "eval_steps_per_second": 4.137, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "eval_accuracy": 0.7522935779816514, | |
| "eval_loss": 0.5918461084365845, | |
| "eval_runtime": 0.9768, | |
| "eval_samples_per_second": 111.587, | |
| "eval_steps_per_second": 4.095, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.7457, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "eval_accuracy": 0.7614678899082569, | |
| "eval_loss": 0.5270918607711792, | |
| "eval_runtime": 0.9695, | |
| "eval_samples_per_second": 112.433, | |
| "eval_steps_per_second": 4.126, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.6834, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "eval_accuracy": 0.944954128440367, | |
| "eval_loss": 0.32956409454345703, | |
| "eval_runtime": 0.9661, | |
| "eval_samples_per_second": 112.825, | |
| "eval_steps_per_second": 4.14, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 3.76984126984127e-05, | |
| "loss": 0.5847, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9174311926605505, | |
| "eval_loss": 0.2882761061191559, | |
| "eval_runtime": 0.9951, | |
| "eval_samples_per_second": 109.538, | |
| "eval_steps_per_second": 4.02, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.5199, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_accuracy": 0.926605504587156, | |
| "eval_loss": 0.28957581520080566, | |
| "eval_runtime": 1.0924, | |
| "eval_samples_per_second": 99.784, | |
| "eval_steps_per_second": 3.662, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "eval_accuracy": 0.8990825688073395, | |
| "eval_loss": 0.28592535853385925, | |
| "eval_runtime": 1.1495, | |
| "eval_samples_per_second": 94.822, | |
| "eval_steps_per_second": 3.48, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 3.3730158730158734e-05, | |
| "loss": 0.4657, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_accuracy": 0.908256880733945, | |
| "eval_loss": 0.2515007257461548, | |
| "eval_runtime": 1.0064, | |
| "eval_samples_per_second": 108.302, | |
| "eval_steps_per_second": 3.974, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.4585, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.908256880733945, | |
| "eval_loss": 0.2260819673538208, | |
| "eval_runtime": 0.9752, | |
| "eval_samples_per_second": 111.775, | |
| "eval_steps_per_second": 4.102, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 16.77, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 0.3892, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "eval_accuracy": 0.926605504587156, | |
| "eval_loss": 0.21420958638191223, | |
| "eval_runtime": 0.9847, | |
| "eval_samples_per_second": 110.697, | |
| "eval_steps_per_second": 4.062, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "eval_accuracy": 0.944954128440367, | |
| "eval_loss": 0.17877857387065887, | |
| "eval_runtime": 0.9683, | |
| "eval_samples_per_second": 112.574, | |
| "eval_steps_per_second": 4.131, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.3939, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "eval_accuracy": 0.926605504587156, | |
| "eval_loss": 0.19475382566452026, | |
| "eval_runtime": 0.9741, | |
| "eval_samples_per_second": 111.895, | |
| "eval_steps_per_second": 4.106, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 2.5793650793650796e-05, | |
| "loss": 0.3429, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.944954128440367, | |
| "eval_loss": 0.16848985850811005, | |
| "eval_runtime": 0.9757, | |
| "eval_samples_per_second": 111.717, | |
| "eval_steps_per_second": 4.1, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 20.65, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.3493, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "eval_accuracy": 0.908256880733945, | |
| "eval_loss": 0.19857284426689148, | |
| "eval_runtime": 0.9817, | |
| "eval_samples_per_second": 111.032, | |
| "eval_steps_per_second": 4.075, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "learning_rate": 2.1825396825396827e-05, | |
| "loss": 0.3462, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "eval_accuracy": 0.9357798165137615, | |
| "eval_loss": 0.15396390855312347, | |
| "eval_runtime": 1.0221, | |
| "eval_samples_per_second": 106.646, | |
| "eval_steps_per_second": 3.914, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "eval_accuracy": 0.944954128440367, | |
| "eval_loss": 0.14491401612758636, | |
| "eval_runtime": 1.1281, | |
| "eval_samples_per_second": 96.619, | |
| "eval_steps_per_second": 3.546, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.3117, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9541284403669725, | |
| "eval_loss": 0.1378978043794632, | |
| "eval_runtime": 1.0483, | |
| "eval_samples_per_second": 103.979, | |
| "eval_steps_per_second": 3.816, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 24.52, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 0.3109, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 24.9, | |
| "eval_accuracy": 0.944954128440367, | |
| "eval_loss": 0.14227117598056793, | |
| "eval_runtime": 0.9868, | |
| "eval_samples_per_second": 110.461, | |
| "eval_steps_per_second": 4.054, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.2867, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 25.94, | |
| "eval_accuracy": 0.944954128440367, | |
| "eval_loss": 0.14505964517593384, | |
| "eval_runtime": 0.9691, | |
| "eval_samples_per_second": 112.475, | |
| "eval_steps_per_second": 4.128, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 26.97, | |
| "eval_accuracy": 0.9724770642201835, | |
| "eval_loss": 0.11537650972604752, | |
| "eval_runtime": 0.9785, | |
| "eval_samples_per_second": 111.4, | |
| "eval_steps_per_second": 4.088, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 27.1, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.293, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9541284403669725, | |
| "eval_loss": 0.11520104110240936, | |
| "eval_runtime": 0.9792, | |
| "eval_samples_per_second": 111.317, | |
| "eval_steps_per_second": 4.085, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 28.39, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 0.2782, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 28.9, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.12612132728099823, | |
| "eval_runtime": 0.9683, | |
| "eval_samples_per_second": 112.567, | |
| "eval_steps_per_second": 4.131, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 29.68, | |
| "learning_rate": 9.92063492063492e-06, | |
| "loss": 0.2744, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 29.94, | |
| "eval_accuracy": 0.981651376146789, | |
| "eval_loss": 0.1174527108669281, | |
| "eval_runtime": 0.973, | |
| "eval_samples_per_second": 112.025, | |
| "eval_steps_per_second": 4.111, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.2711, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.1291634738445282, | |
| "eval_runtime": 0.9731, | |
| "eval_samples_per_second": 112.011, | |
| "eval_steps_per_second": 4.11, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.981651376146789, | |
| "eval_loss": 0.11013606935739517, | |
| "eval_runtime": 1.0636, | |
| "eval_samples_per_second": 102.48, | |
| "eval_steps_per_second": 3.761, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 0.2652, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 32.9, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.12023505568504333, | |
| "eval_runtime": 1.1435, | |
| "eval_samples_per_second": 95.321, | |
| "eval_steps_per_second": 3.498, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 33.55, | |
| "learning_rate": 3.968253968253968e-06, | |
| "loss": 0.2218, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 33.94, | |
| "eval_accuracy": 0.981651376146789, | |
| "eval_loss": 0.1118844598531723, | |
| "eval_runtime": 0.9784, | |
| "eval_samples_per_second": 111.404, | |
| "eval_steps_per_second": 4.088, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 34.84, | |
| "learning_rate": 1.984126984126984e-06, | |
| "loss": 0.2899, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 34.97, | |
| "eval_accuracy": 0.981651376146789, | |
| "eval_loss": 0.10707387328147888, | |
| "eval_runtime": 0.9787, | |
| "eval_samples_per_second": 111.376, | |
| "eval_steps_per_second": 4.087, | |
| "step": 271 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.981651376146789, | |
| "eval_loss": 0.1077311784029007, | |
| "eval_runtime": 0.9716, | |
| "eval_samples_per_second": 112.188, | |
| "eval_steps_per_second": 4.117, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "learning_rate": 0.0, | |
| "loss": 0.2143, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "eval_accuracy": 0.981651376146789, | |
| "eval_loss": 0.10772820562124252, | |
| "eval_runtime": 1.1396, | |
| "eval_samples_per_second": 95.65, | |
| "eval_steps_per_second": 3.51, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "step": 280, | |
| "total_flos": 1.1496971422295654e+18, | |
| "train_loss": 0.538695216178894, | |
| "train_runtime": 832.5003, | |
| "train_samples_per_second": 46.991, | |
| "train_steps_per_second": 0.336 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 280, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 1.1496971422295654e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |