| { | |
| "best_metric": 0.9716738197424892, | |
| "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-lora-medmnistv2/checkpoint-382", | |
| "epoch": 9.882352941176471, | |
| "eval_steps": 500, | |
| "global_step": 630, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 1.3673616647720337, | |
| "learning_rate": 0.004920634920634921, | |
| "loss": 0.4969, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.564309298992157, | |
| "learning_rate": 0.004841269841269842, | |
| "loss": 0.3486, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.83164381980896, | |
| "learning_rate": 0.0047619047619047615, | |
| "loss": 0.2505, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.7283287644386292, | |
| "learning_rate": 0.004682539682539683, | |
| "loss": 0.2221, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.3768545687198639, | |
| "learning_rate": 0.004603174603174603, | |
| "loss": 0.2593, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.4381789267063141, | |
| "learning_rate": 0.004523809523809524, | |
| "loss": 0.211, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.9605150214592275, | |
| "eval_f1": 0.9500622474690998, | |
| "eval_loss": 0.11396056413650513, | |
| "eval_precision": 0.9400752542471662, | |
| "eval_recall": 0.9615826914848096, | |
| "eval_runtime": 11.8657, | |
| "eval_samples_per_second": 98.182, | |
| "eval_steps_per_second": 6.152, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.8330244421958923, | |
| "learning_rate": 0.0044444444444444444, | |
| "loss": 0.2013, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.0441814661026, | |
| "learning_rate": 0.004365079365079365, | |
| "loss": 0.1882, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 0.8081804513931274, | |
| "learning_rate": 0.004285714285714286, | |
| "loss": 0.2045, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "grad_norm": 0.49920302629470825, | |
| "learning_rate": 0.004206349206349207, | |
| "loss": 0.194, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 0.39656010270118713, | |
| "learning_rate": 0.0041269841269841265, | |
| "loss": 0.1918, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.4606592655181885, | |
| "learning_rate": 0.004047619047619048, | |
| "loss": 0.1911, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.9330472103004291, | |
| "eval_f1": 0.9185857738786393, | |
| "eval_loss": 0.151698499917984, | |
| "eval_precision": 0.8989291177970423, | |
| "eval_recall": 0.948319732257473, | |
| "eval_runtime": 11.9483, | |
| "eval_samples_per_second": 97.504, | |
| "eval_steps_per_second": 6.11, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 0.6359087824821472, | |
| "learning_rate": 0.003976190476190476, | |
| "loss": 0.1652, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.439413845539093, | |
| "learning_rate": 0.003896825396825397, | |
| "loss": 0.1795, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 0.9134456515312195, | |
| "learning_rate": 0.003817460317460317, | |
| "loss": 0.2155, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "grad_norm": 1.381035566329956, | |
| "learning_rate": 0.0037380952380952383, | |
| "loss": 0.218, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 0.7177550196647644, | |
| "learning_rate": 0.0036587301587301586, | |
| "loss": 0.1839, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "grad_norm": 0.7977726459503174, | |
| "learning_rate": 0.0035793650793650793, | |
| "loss": 0.1595, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "grad_norm": 1.0747120380401611, | |
| "learning_rate": 0.0034999999999999996, | |
| "loss": 0.1695, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9579399141630901, | |
| "eval_f1": 0.9470671089099915, | |
| "eval_loss": 0.11627380549907684, | |
| "eval_precision": 0.9354318260568261, | |
| "eval_recall": 0.960904547955254, | |
| "eval_runtime": 11.7288, | |
| "eval_samples_per_second": 99.328, | |
| "eval_steps_per_second": 6.224, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "grad_norm": 0.328319787979126, | |
| "learning_rate": 0.003420634920634921, | |
| "loss": 0.2, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "grad_norm": 0.33479979634284973, | |
| "learning_rate": 0.003341269841269841, | |
| "loss": 0.1086, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "grad_norm": 0.2731302082538605, | |
| "learning_rate": 0.003261904761904762, | |
| "loss": 0.1782, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "grad_norm": 0.507108211517334, | |
| "learning_rate": 0.0031825396825396826, | |
| "loss": 0.1869, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "grad_norm": 2.0676915645599365, | |
| "learning_rate": 0.0031031746031746034, | |
| "loss": 0.1438, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "grad_norm": 0.9014037847518921, | |
| "learning_rate": 0.0030317460317460317, | |
| "loss": 0.1556, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9570815450643777, | |
| "eval_f1": 0.9416580530916725, | |
| "eval_loss": 0.11594364047050476, | |
| "eval_precision": 0.9668957588361103, | |
| "eval_recall": 0.9220192248914971, | |
| "eval_runtime": 11.9006, | |
| "eval_samples_per_second": 97.894, | |
| "eval_steps_per_second": 6.134, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 0.5203756093978882, | |
| "learning_rate": 0.0029523809523809524, | |
| "loss": 0.1301, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "grad_norm": 0.2871193587779999, | |
| "learning_rate": 0.0028730158730158727, | |
| "loss": 0.1614, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "grad_norm": 0.5260260701179504, | |
| "learning_rate": 0.002793650793650794, | |
| "loss": 0.1536, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "grad_norm": 1.3978557586669922, | |
| "learning_rate": 0.0027142857142857142, | |
| "loss": 0.1558, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "grad_norm": 0.3816111385822296, | |
| "learning_rate": 0.002634920634920635, | |
| "loss": 0.1393, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "grad_norm": 0.8400008678436279, | |
| "learning_rate": 0.0025555555555555553, | |
| "loss": 0.173, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.9502145922746781, | |
| "eval_f1": 0.9381227106227107, | |
| "eval_loss": 0.11663959920406342, | |
| "eval_precision": 0.9228616598558795, | |
| "eval_recall": 0.9578061006173972, | |
| "eval_runtime": 11.8429, | |
| "eval_samples_per_second": 98.371, | |
| "eval_steps_per_second": 6.164, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "grad_norm": 0.4521535336971283, | |
| "learning_rate": 0.0024761904761904764, | |
| "loss": 0.1657, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "grad_norm": 1.030579924583435, | |
| "learning_rate": 0.0024047619047619048, | |
| "loss": 0.1658, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "grad_norm": 0.7163300514221191, | |
| "learning_rate": 0.0023253968253968255, | |
| "loss": 0.142, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "grad_norm": 0.45034098625183105, | |
| "learning_rate": 0.0022460317460317463, | |
| "loss": 0.1319, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "grad_norm": 0.8648952841758728, | |
| "learning_rate": 0.002166666666666667, | |
| "loss": 0.1621, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "grad_norm": 1.5358220338821411, | |
| "learning_rate": 0.0020873015873015873, | |
| "loss": 0.1747, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "grad_norm": 4.473792552947998, | |
| "learning_rate": 0.002007936507936508, | |
| "loss": 0.1485, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "eval_accuracy": 0.9716738197424892, | |
| "eval_f1": 0.9637751828201366, | |
| "eval_loss": 0.08252137899398804, | |
| "eval_precision": 0.9578261710612463, | |
| "eval_recall": 0.9701960694418974, | |
| "eval_runtime": 12.3059, | |
| "eval_samples_per_second": 94.67, | |
| "eval_steps_per_second": 5.932, | |
| "step": 382 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "grad_norm": 0.918424665927887, | |
| "learning_rate": 0.0019285714285714288, | |
| "loss": 0.1477, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "grad_norm": 0.2846250534057617, | |
| "learning_rate": 0.0018492063492063493, | |
| "loss": 0.107, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "grad_norm": 1.0383787155151367, | |
| "learning_rate": 0.00176984126984127, | |
| "loss": 0.1429, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 6.59, | |
| "grad_norm": 0.5193153619766235, | |
| "learning_rate": 0.0016904761904761906, | |
| "loss": 0.1611, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "grad_norm": 0.6372103095054626, | |
| "learning_rate": 0.0016111111111111113, | |
| "loss": 0.1602, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "grad_norm": 0.4708787202835083, | |
| "learning_rate": 0.0015317460317460319, | |
| "loss": 0.1854, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9716738197424892, | |
| "eval_f1": 0.9637751828201366, | |
| "eval_loss": 0.08776707202196121, | |
| "eval_precision": 0.9578261710612463, | |
| "eval_recall": 0.9701960694418974, | |
| "eval_runtime": 11.6697, | |
| "eval_samples_per_second": 99.831, | |
| "eval_steps_per_second": 6.256, | |
| "step": 446 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "grad_norm": 1.3629688024520874, | |
| "learning_rate": 0.0014523809523809526, | |
| "loss": 0.2118, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "grad_norm": 0.5590702295303345, | |
| "learning_rate": 0.0013730158730158731, | |
| "loss": 0.1608, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "grad_norm": 1.0309607982635498, | |
| "learning_rate": 0.0012936507936507939, | |
| "loss": 0.1684, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "grad_norm": 0.41553324460983276, | |
| "learning_rate": 0.0012142857142857144, | |
| "loss": 0.116, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "grad_norm": 0.3408653736114502, | |
| "learning_rate": 0.001134920634920635, | |
| "loss": 0.1188, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "grad_norm": 0.4753941595554352, | |
| "learning_rate": 0.0010555555555555557, | |
| "loss": 0.1345, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.5954430103302002, | |
| "learning_rate": 0.0009761904761904762, | |
| "loss": 0.1353, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9587982832618026, | |
| "eval_f1": 0.9483988129512335, | |
| "eval_loss": 0.10603910684585571, | |
| "eval_precision": 0.9351169479762169, | |
| "eval_recall": 0.9646773182957393, | |
| "eval_runtime": 11.7084, | |
| "eval_samples_per_second": 99.501, | |
| "eval_steps_per_second": 6.235, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "grad_norm": 0.35974177718162537, | |
| "learning_rate": 0.0008968253968253968, | |
| "loss": 0.1192, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 8.31, | |
| "grad_norm": 0.4420110285282135, | |
| "learning_rate": 0.0008174603174603175, | |
| "loss": 0.1342, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "grad_norm": 0.83766108751297, | |
| "learning_rate": 0.0007380952380952381, | |
| "loss": 0.1206, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "grad_norm": 0.488089919090271, | |
| "learning_rate": 0.0006587301587301587, | |
| "loss": 0.1288, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "grad_norm": 0.48352861404418945, | |
| "learning_rate": 0.0005793650793650794, | |
| "loss": 0.1376, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "grad_norm": 0.504624605178833, | |
| "learning_rate": 0.0005, | |
| "loss": 0.1196, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.9690987124463519, | |
| "eval_f1": 0.9606826261080863, | |
| "eval_loss": 0.08817815780639648, | |
| "eval_precision": 0.9527364558696758, | |
| "eval_recall": 0.9695179259123419, | |
| "eval_runtime": 11.6769, | |
| "eval_samples_per_second": 99.77, | |
| "eval_steps_per_second": 6.252, | |
| "step": 573 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "grad_norm": 0.22534619271755219, | |
| "learning_rate": 0.0004206349206349207, | |
| "loss": 0.1083, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "grad_norm": 0.642715573310852, | |
| "learning_rate": 0.0003412698412698412, | |
| "loss": 0.095, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "grad_norm": 0.5458790063858032, | |
| "learning_rate": 0.0002619047619047619, | |
| "loss": 0.1413, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "grad_norm": 0.44277364015579224, | |
| "learning_rate": 0.00018253968253968255, | |
| "loss": 0.1277, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "grad_norm": 0.5391418933868408, | |
| "learning_rate": 0.00010317460317460317, | |
| "loss": 0.1289, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "grad_norm": 0.3602410554885864, | |
| "learning_rate": 2.3809523809523814e-05, | |
| "loss": 0.1218, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "eval_accuracy": 0.9639484978540772, | |
| "eval_f1": 0.9547614644970415, | |
| "eval_loss": 0.09816381335258484, | |
| "eval_precision": 0.9419247946464473, | |
| "eval_recall": 0.9702896723516108, | |
| "eval_runtime": 11.5253, | |
| "eval_samples_per_second": 101.082, | |
| "eval_steps_per_second": 6.334, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "step": 630, | |
| "total_flos": 3.1439051980091965e+18, | |
| "train_loss": 0.16921042270130582, | |
| "train_runtime": 727.4236, | |
| "train_samples_per_second": 56.047, | |
| "train_steps_per_second": 0.866 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 630, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 3.1439051980091965e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |