|
{ |
|
"best_metric": 2.827199935913086, |
|
"best_model_checkpoint": "cdetr-mist1-brain-gt-tumors-8ah-6l/checkpoint-4830", |
|
"epoch": 50.0, |
|
"eval_steps": 500, |
|
"global_step": 5750, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 9.624347826086956e-06, |
|
"loss": 5.4149, |
|
"step": 115 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 4.397365093231201, |
|
"eval_runtime": 5.8932, |
|
"eval_samples_per_second": 6.787, |
|
"eval_steps_per_second": 0.848, |
|
"step": 115 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 9.231304347826087e-06, |
|
"loss": 3.9453, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_loss": 3.651952028274536, |
|
"eval_runtime": 5.9289, |
|
"eval_samples_per_second": 6.747, |
|
"eval_steps_per_second": 0.843, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 8.831304347826088e-06, |
|
"loss": 3.7269, |
|
"step": 345 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_loss": 3.7602438926696777, |
|
"eval_runtime": 5.8799, |
|
"eval_samples_per_second": 6.803, |
|
"eval_steps_per_second": 0.85, |
|
"step": 345 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 8.434782608695653e-06, |
|
"loss": 3.5898, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_loss": 3.5670828819274902, |
|
"eval_runtime": 5.9645, |
|
"eval_samples_per_second": 6.706, |
|
"eval_steps_per_second": 0.838, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 8.034782608695652e-06, |
|
"loss": 3.486, |
|
"step": 575 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_loss": 3.4911675453186035, |
|
"eval_runtime": 5.9241, |
|
"eval_samples_per_second": 6.752, |
|
"eval_steps_per_second": 0.844, |
|
"step": 575 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"learning_rate": 7.634782608695653e-06, |
|
"loss": 3.4073, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_loss": 3.409463405609131, |
|
"eval_runtime": 5.8302, |
|
"eval_samples_per_second": 6.861, |
|
"eval_steps_per_second": 0.858, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"learning_rate": 7.234782608695653e-06, |
|
"loss": 3.4181, |
|
"step": 805 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_loss": 3.3183083534240723, |
|
"eval_runtime": 5.8894, |
|
"eval_samples_per_second": 6.792, |
|
"eval_steps_per_second": 0.849, |
|
"step": 805 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"learning_rate": 6.834782608695653e-06, |
|
"loss": 3.3603, |
|
"step": 920 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_loss": 3.1111273765563965, |
|
"eval_runtime": 5.9175, |
|
"eval_samples_per_second": 6.76, |
|
"eval_steps_per_second": 0.845, |
|
"step": 920 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"learning_rate": 6.434782608695652e-06, |
|
"loss": 3.2777, |
|
"step": 1035 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_loss": 3.1992499828338623, |
|
"eval_runtime": 5.9275, |
|
"eval_samples_per_second": 6.748, |
|
"eval_steps_per_second": 0.844, |
|
"step": 1035 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"learning_rate": 6.034782608695653e-06, |
|
"loss": 3.2851, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_loss": 3.3997130393981934, |
|
"eval_runtime": 5.8939, |
|
"eval_samples_per_second": 6.787, |
|
"eval_steps_per_second": 0.848, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"learning_rate": 5.634782608695652e-06, |
|
"loss": 3.266, |
|
"step": 1265 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"eval_loss": 3.286081314086914, |
|
"eval_runtime": 6.1287, |
|
"eval_samples_per_second": 6.527, |
|
"eval_steps_per_second": 0.816, |
|
"step": 1265 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"learning_rate": 5.234782608695652e-06, |
|
"loss": 3.2803, |
|
"step": 1380 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_loss": 3.18131947517395, |
|
"eval_runtime": 6.1123, |
|
"eval_samples_per_second": 6.544, |
|
"eval_steps_per_second": 0.818, |
|
"step": 1380 |
|
}, |
|
{ |
|
"epoch": 13.0, |
|
"learning_rate": 4.834782608695652e-06, |
|
"loss": 3.1733, |
|
"step": 1495 |
|
}, |
|
{ |
|
"epoch": 13.0, |
|
"eval_loss": 2.9837594032287598, |
|
"eval_runtime": 5.925, |
|
"eval_samples_per_second": 6.751, |
|
"eval_steps_per_second": 0.844, |
|
"step": 1495 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"learning_rate": 4.434782608695653e-06, |
|
"loss": 3.2094, |
|
"step": 1610 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"eval_loss": 3.1175174713134766, |
|
"eval_runtime": 5.9137, |
|
"eval_samples_per_second": 6.764, |
|
"eval_steps_per_second": 0.845, |
|
"step": 1610 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"learning_rate": 4.034782608695652e-06, |
|
"loss": 3.1718, |
|
"step": 1725 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"eval_loss": 3.0063586235046387, |
|
"eval_runtime": 5.9077, |
|
"eval_samples_per_second": 6.771, |
|
"eval_steps_per_second": 0.846, |
|
"step": 1725 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"learning_rate": 3.6347826086956524e-06, |
|
"loss": 3.1303, |
|
"step": 1840 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"eval_loss": 3.086920738220215, |
|
"eval_runtime": 5.9305, |
|
"eval_samples_per_second": 6.745, |
|
"eval_steps_per_second": 0.843, |
|
"step": 1840 |
|
}, |
|
{ |
|
"epoch": 17.0, |
|
"learning_rate": 3.2347826086956524e-06, |
|
"loss": 3.0897, |
|
"step": 1955 |
|
}, |
|
{ |
|
"epoch": 17.0, |
|
"eval_loss": 3.0305533409118652, |
|
"eval_runtime": 6.0975, |
|
"eval_samples_per_second": 6.56, |
|
"eval_steps_per_second": 0.82, |
|
"step": 1955 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"learning_rate": 2.8347826086956525e-06, |
|
"loss": 3.0233, |
|
"step": 2070 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"eval_loss": 2.9478588104248047, |
|
"eval_runtime": 5.9801, |
|
"eval_samples_per_second": 6.689, |
|
"eval_steps_per_second": 0.836, |
|
"step": 2070 |
|
}, |
|
{ |
|
"epoch": 19.0, |
|
"learning_rate": 2.4347826086956525e-06, |
|
"loss": 3.0156, |
|
"step": 2185 |
|
}, |
|
{ |
|
"epoch": 19.0, |
|
"eval_loss": 2.914536952972412, |
|
"eval_runtime": 6.0155, |
|
"eval_samples_per_second": 6.649, |
|
"eval_steps_per_second": 0.831, |
|
"step": 2185 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"learning_rate": 2.0347826086956525e-06, |
|
"loss": 3.0277, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"eval_loss": 2.891925811767578, |
|
"eval_runtime": 6.058, |
|
"eval_samples_per_second": 6.603, |
|
"eval_steps_per_second": 0.825, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 21.0, |
|
"learning_rate": 1.6382608695652176e-06, |
|
"loss": 3.0847, |
|
"step": 2415 |
|
}, |
|
{ |
|
"epoch": 21.0, |
|
"eval_loss": 2.9321396350860596, |
|
"eval_runtime": 6.1412, |
|
"eval_samples_per_second": 6.513, |
|
"eval_steps_per_second": 0.814, |
|
"step": 2415 |
|
}, |
|
{ |
|
"epoch": 22.0, |
|
"learning_rate": 1.2382608695652176e-06, |
|
"loss": 3.0333, |
|
"step": 2530 |
|
}, |
|
{ |
|
"epoch": 22.0, |
|
"eval_loss": 2.91283917427063, |
|
"eval_runtime": 6.139, |
|
"eval_samples_per_second": 6.516, |
|
"eval_steps_per_second": 0.814, |
|
"step": 2530 |
|
}, |
|
{ |
|
"epoch": 23.0, |
|
"learning_rate": 8.382608695652175e-07, |
|
"loss": 3.0126, |
|
"step": 2645 |
|
}, |
|
{ |
|
"epoch": 23.0, |
|
"eval_loss": 2.8627383708953857, |
|
"eval_runtime": 6.0565, |
|
"eval_samples_per_second": 6.604, |
|
"eval_steps_per_second": 0.826, |
|
"step": 2645 |
|
}, |
|
{ |
|
"epoch": 24.0, |
|
"learning_rate": 5.234782608695652e-06, |
|
"loss": 2.9968, |
|
"step": 2760 |
|
}, |
|
{ |
|
"epoch": 24.0, |
|
"eval_loss": 3.0186123847961426, |
|
"eval_runtime": 5.7749, |
|
"eval_samples_per_second": 6.926, |
|
"eval_steps_per_second": 0.866, |
|
"step": 2760 |
|
}, |
|
{ |
|
"epoch": 25.0, |
|
"learning_rate": 5.036521739130436e-06, |
|
"loss": 3.0295, |
|
"step": 2875 |
|
}, |
|
{ |
|
"epoch": 25.0, |
|
"eval_loss": 3.0147616863250732, |
|
"eval_runtime": 5.799, |
|
"eval_samples_per_second": 6.898, |
|
"eval_steps_per_second": 0.862, |
|
"step": 2875 |
|
}, |
|
{ |
|
"epoch": 26.0, |
|
"learning_rate": 4.836521739130435e-06, |
|
"loss": 3.0294, |
|
"step": 2990 |
|
}, |
|
{ |
|
"epoch": 26.0, |
|
"eval_loss": 3.034066915512085, |
|
"eval_runtime": 5.8474, |
|
"eval_samples_per_second": 6.841, |
|
"eval_steps_per_second": 0.855, |
|
"step": 2990 |
|
}, |
|
{ |
|
"epoch": 27.0, |
|
"learning_rate": 4.636521739130435e-06, |
|
"loss": 3.0395, |
|
"step": 3105 |
|
}, |
|
{ |
|
"epoch": 27.0, |
|
"eval_loss": 2.99969220161438, |
|
"eval_runtime": 5.7579, |
|
"eval_samples_per_second": 6.947, |
|
"eval_steps_per_second": 0.868, |
|
"step": 3105 |
|
}, |
|
{ |
|
"epoch": 28.0, |
|
"learning_rate": 4.436521739130435e-06, |
|
"loss": 3.0445, |
|
"step": 3220 |
|
}, |
|
{ |
|
"epoch": 28.0, |
|
"eval_loss": 3.0574753284454346, |
|
"eval_runtime": 5.6888, |
|
"eval_samples_per_second": 7.031, |
|
"eval_steps_per_second": 0.879, |
|
"step": 3220 |
|
}, |
|
{ |
|
"epoch": 29.0, |
|
"learning_rate": 4.236521739130436e-06, |
|
"loss": 2.9761, |
|
"step": 3335 |
|
}, |
|
{ |
|
"epoch": 29.0, |
|
"eval_loss": 2.97066068649292, |
|
"eval_runtime": 5.7458, |
|
"eval_samples_per_second": 6.962, |
|
"eval_steps_per_second": 0.87, |
|
"step": 3335 |
|
}, |
|
{ |
|
"epoch": 30.0, |
|
"learning_rate": 4.038260869565218e-06, |
|
"loss": 3.0075, |
|
"step": 3450 |
|
}, |
|
{ |
|
"epoch": 30.0, |
|
"eval_loss": 2.939246416091919, |
|
"eval_runtime": 5.722, |
|
"eval_samples_per_second": 6.991, |
|
"eval_steps_per_second": 0.874, |
|
"step": 3450 |
|
}, |
|
{ |
|
"epoch": 31.0, |
|
"learning_rate": 3.838260869565218e-06, |
|
"loss": 3.0198, |
|
"step": 3565 |
|
}, |
|
{ |
|
"epoch": 31.0, |
|
"eval_loss": 2.9121549129486084, |
|
"eval_runtime": 5.8674, |
|
"eval_samples_per_second": 6.817, |
|
"eval_steps_per_second": 0.852, |
|
"step": 3565 |
|
}, |
|
{ |
|
"epoch": 32.0, |
|
"learning_rate": 3.6382608695652177e-06, |
|
"loss": 2.9782, |
|
"step": 3680 |
|
}, |
|
{ |
|
"epoch": 32.0, |
|
"eval_loss": 2.947079658508301, |
|
"eval_runtime": 5.7346, |
|
"eval_samples_per_second": 6.975, |
|
"eval_steps_per_second": 0.872, |
|
"step": 3680 |
|
}, |
|
{ |
|
"epoch": 33.0, |
|
"learning_rate": 3.4382608695652175e-06, |
|
"loss": 2.9773, |
|
"step": 3795 |
|
}, |
|
{ |
|
"epoch": 33.0, |
|
"eval_loss": 3.030561923980713, |
|
"eval_runtime": 5.7197, |
|
"eval_samples_per_second": 6.993, |
|
"eval_steps_per_second": 0.874, |
|
"step": 3795 |
|
}, |
|
{ |
|
"epoch": 34.0, |
|
"learning_rate": 3.2382608695652177e-06, |
|
"loss": 2.9528, |
|
"step": 3910 |
|
}, |
|
{ |
|
"epoch": 34.0, |
|
"eval_loss": 2.8513386249542236, |
|
"eval_runtime": 5.6599, |
|
"eval_samples_per_second": 7.067, |
|
"eval_steps_per_second": 0.883, |
|
"step": 3910 |
|
}, |
|
{ |
|
"epoch": 35.0, |
|
"learning_rate": 3.038260869565218e-06, |
|
"loss": 2.9228, |
|
"step": 4025 |
|
}, |
|
{ |
|
"epoch": 35.0, |
|
"eval_loss": 2.899709701538086, |
|
"eval_runtime": 5.6812, |
|
"eval_samples_per_second": 7.041, |
|
"eval_steps_per_second": 0.88, |
|
"step": 4025 |
|
}, |
|
{ |
|
"epoch": 36.0, |
|
"learning_rate": 2.8382608695652177e-06, |
|
"loss": 2.9221, |
|
"step": 4140 |
|
}, |
|
{ |
|
"epoch": 36.0, |
|
"eval_loss": 2.8646340370178223, |
|
"eval_runtime": 5.6561, |
|
"eval_samples_per_second": 7.072, |
|
"eval_steps_per_second": 0.884, |
|
"step": 4140 |
|
}, |
|
{ |
|
"epoch": 37.0, |
|
"learning_rate": 2.638260869565218e-06, |
|
"loss": 2.8933, |
|
"step": 4255 |
|
}, |
|
{ |
|
"epoch": 37.0, |
|
"eval_loss": 2.8870697021484375, |
|
"eval_runtime": 5.7007, |
|
"eval_samples_per_second": 7.017, |
|
"eval_steps_per_second": 0.877, |
|
"step": 4255 |
|
}, |
|
{ |
|
"epoch": 38.0, |
|
"learning_rate": 2.4382608695652177e-06, |
|
"loss": 2.8925, |
|
"step": 4370 |
|
}, |
|
{ |
|
"epoch": 38.0, |
|
"eval_loss": 2.94067120552063, |
|
"eval_runtime": 5.6567, |
|
"eval_samples_per_second": 7.071, |
|
"eval_steps_per_second": 0.884, |
|
"step": 4370 |
|
}, |
|
{ |
|
"epoch": 39.0, |
|
"learning_rate": 2.2382608695652175e-06, |
|
"loss": 2.9069, |
|
"step": 4485 |
|
}, |
|
{ |
|
"epoch": 39.0, |
|
"eval_loss": 2.9625325202941895, |
|
"eval_runtime": 5.6615, |
|
"eval_samples_per_second": 7.065, |
|
"eval_steps_per_second": 0.883, |
|
"step": 4485 |
|
}, |
|
{ |
|
"epoch": 40.0, |
|
"learning_rate": 2.0382608695652177e-06, |
|
"loss": 2.9246, |
|
"step": 4600 |
|
}, |
|
{ |
|
"epoch": 40.0, |
|
"eval_loss": 2.994633197784424, |
|
"eval_runtime": 5.6894, |
|
"eval_samples_per_second": 7.031, |
|
"eval_steps_per_second": 0.879, |
|
"step": 4600 |
|
}, |
|
{ |
|
"epoch": 41.0, |
|
"learning_rate": 1.8382608695652178e-06, |
|
"loss": 2.9089, |
|
"step": 4715 |
|
}, |
|
{ |
|
"epoch": 41.0, |
|
"eval_loss": 2.8935651779174805, |
|
"eval_runtime": 5.7226, |
|
"eval_samples_per_second": 6.99, |
|
"eval_steps_per_second": 0.874, |
|
"step": 4715 |
|
}, |
|
{ |
|
"epoch": 42.0, |
|
"learning_rate": 1.6382608695652176e-06, |
|
"loss": 2.8573, |
|
"step": 4830 |
|
}, |
|
{ |
|
"epoch": 42.0, |
|
"eval_loss": 2.827199935913086, |
|
"eval_runtime": 5.6411, |
|
"eval_samples_per_second": 7.091, |
|
"eval_steps_per_second": 0.886, |
|
"step": 4830 |
|
}, |
|
{ |
|
"epoch": 43.0, |
|
"learning_rate": 1.4382608695652176e-06, |
|
"loss": 2.8378, |
|
"step": 4945 |
|
}, |
|
{ |
|
"epoch": 43.0, |
|
"eval_loss": 2.8543009757995605, |
|
"eval_runtime": 5.6643, |
|
"eval_samples_per_second": 7.062, |
|
"eval_steps_per_second": 0.883, |
|
"step": 4945 |
|
}, |
|
{ |
|
"epoch": 44.0, |
|
"learning_rate": 1.2382608695652176e-06, |
|
"loss": 2.8957, |
|
"step": 5060 |
|
}, |
|
{ |
|
"epoch": 44.0, |
|
"eval_loss": 2.859046459197998, |
|
"eval_runtime": 5.687, |
|
"eval_samples_per_second": 7.034, |
|
"eval_steps_per_second": 0.879, |
|
"step": 5060 |
|
}, |
|
{ |
|
"epoch": 45.0, |
|
"learning_rate": 1.0382608695652174e-06, |
|
"loss": 2.8454, |
|
"step": 5175 |
|
}, |
|
{ |
|
"epoch": 45.0, |
|
"eval_loss": 2.8524787425994873, |
|
"eval_runtime": 5.6623, |
|
"eval_samples_per_second": 7.064, |
|
"eval_steps_per_second": 0.883, |
|
"step": 5175 |
|
}, |
|
{ |
|
"epoch": 46.0, |
|
"learning_rate": 8.382608695652175e-07, |
|
"loss": 2.7964, |
|
"step": 5290 |
|
}, |
|
{ |
|
"epoch": 46.0, |
|
"eval_loss": 2.8657748699188232, |
|
"eval_runtime": 5.6625, |
|
"eval_samples_per_second": 7.064, |
|
"eval_steps_per_second": 0.883, |
|
"step": 5290 |
|
}, |
|
{ |
|
"epoch": 47.0, |
|
"learning_rate": 6.382608695652175e-07, |
|
"loss": 2.8172, |
|
"step": 5405 |
|
}, |
|
{ |
|
"epoch": 47.0, |
|
"eval_loss": 2.8872039318084717, |
|
"eval_runtime": 5.6798, |
|
"eval_samples_per_second": 7.043, |
|
"eval_steps_per_second": 0.88, |
|
"step": 5405 |
|
}, |
|
{ |
|
"epoch": 48.0, |
|
"learning_rate": 4.382608695652174e-07, |
|
"loss": 2.8472, |
|
"step": 5520 |
|
}, |
|
{ |
|
"epoch": 48.0, |
|
"eval_loss": 2.8653695583343506, |
|
"eval_runtime": 5.6595, |
|
"eval_samples_per_second": 7.068, |
|
"eval_steps_per_second": 0.883, |
|
"step": 5520 |
|
}, |
|
{ |
|
"epoch": 49.0, |
|
"learning_rate": 2.3826086956521743e-07, |
|
"loss": 2.811, |
|
"step": 5635 |
|
}, |
|
{ |
|
"epoch": 49.0, |
|
"eval_loss": 2.8571643829345703, |
|
"eval_runtime": 5.7272, |
|
"eval_samples_per_second": 6.984, |
|
"eval_steps_per_second": 0.873, |
|
"step": 5635 |
|
}, |
|
{ |
|
"epoch": 50.0, |
|
"learning_rate": 3.82608695652174e-08, |
|
"loss": 2.801, |
|
"step": 5750 |
|
}, |
|
{ |
|
"epoch": 50.0, |
|
"eval_loss": 2.830277919769287, |
|
"eval_runtime": 5.6412, |
|
"eval_samples_per_second": 7.091, |
|
"eval_steps_per_second": 0.886, |
|
"step": 5750 |
|
}, |
|
{ |
|
"epoch": 50.0, |
|
"step": 5750, |
|
"total_flos": 1.147766207616e+19, |
|
"train_loss": 1.5786318518597147, |
|
"train_runtime": 2642.836, |
|
"train_samples_per_second": 8.703, |
|
"train_steps_per_second": 2.176 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 5750, |
|
"num_train_epochs": 50, |
|
"save_steps": 500, |
|
"total_flos": 1.147766207616e+19, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|