| { | |
| "best_metric": 0.975107296137339, | |
| "best_model_checkpoint": "deit-base-patch16-224-finetuned-lora-medmnistv2/checkpoint-446", | |
| "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": 1.1300164461135864, | |
| "learning_rate": 0.004920634920634921, | |
| "loss": 0.4737, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.7213650941848755, | |
| "learning_rate": 0.004841269841269842, | |
| "loss": 0.6061, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 3.0624804496765137, | |
| "learning_rate": 0.0047619047619047615, | |
| "loss": 0.2754, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.509834885597229, | |
| "learning_rate": 0.004682539682539683, | |
| "loss": 0.2462, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 1.2371031045913696, | |
| "learning_rate": 0.004603174603174603, | |
| "loss": 0.205, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 1.2106962203979492, | |
| "learning_rate": 0.004523809523809524, | |
| "loss": 0.2556, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.9184549356223176, | |
| "eval_f1": 0.8822492390220755, | |
| "eval_loss": 0.20187422633171082, | |
| "eval_precision": 0.9446022031823746, | |
| "eval_recall": 0.8469420502475702, | |
| "eval_runtime": 11.2928, | |
| "eval_samples_per_second": 103.163, | |
| "eval_steps_per_second": 6.464, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 2.5079874992370605, | |
| "learning_rate": 0.0044444444444444444, | |
| "loss": 0.282, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.6824792623519897, | |
| "learning_rate": 0.004365079365079365, | |
| "loss": 0.2291, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 1.294320821762085, | |
| "learning_rate": 0.004285714285714286, | |
| "loss": 0.208, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "grad_norm": 2.1044976711273193, | |
| "learning_rate": 0.004206349206349207, | |
| "loss": 0.2288, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 0.8730837106704712, | |
| "learning_rate": 0.0041269841269841265, | |
| "loss": 0.3071, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 1.6219590902328491, | |
| "learning_rate": 0.004047619047619048, | |
| "loss": 0.2302, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.9613733905579399, | |
| "eval_f1": 0.9513881612438695, | |
| "eval_loss": 0.10982219129800797, | |
| "eval_precision": 0.9396367521367521, | |
| "eval_recall": 0.965355461825295, | |
| "eval_runtime": 11.2742, | |
| "eval_samples_per_second": 103.333, | |
| "eval_steps_per_second": 6.475, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 1.7205462455749512, | |
| "learning_rate": 0.003968253968253968, | |
| "loss": 0.214, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.4880439341068268, | |
| "learning_rate": 0.003888888888888889, | |
| "loss": 0.2757, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 1.5420565605163574, | |
| "learning_rate": 0.0038095238095238095, | |
| "loss": 0.2314, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "grad_norm": 0.872795820236206, | |
| "learning_rate": 0.0037301587301587303, | |
| "loss": 0.2689, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 1.3670737743377686, | |
| "learning_rate": 0.0036507936507936506, | |
| "loss": 0.1787, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "grad_norm": 0.9697684049606323, | |
| "learning_rate": 0.0035714285714285718, | |
| "loss": 0.1962, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "grad_norm": 0.6434329152107239, | |
| "learning_rate": 0.003492063492063492, | |
| "loss": 0.2258, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9622317596566523, | |
| "eval_f1": 0.9496135003577972, | |
| "eval_loss": 0.11506476998329163, | |
| "eval_precision": 0.9640835246121029, | |
| "eval_recall": 0.9372077296900788, | |
| "eval_runtime": 11.448, | |
| "eval_samples_per_second": 101.764, | |
| "eval_steps_per_second": 6.377, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "grad_norm": 0.3758108913898468, | |
| "learning_rate": 0.003412698412698413, | |
| "loss": 0.2048, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "grad_norm": 0.3552361726760864, | |
| "learning_rate": 0.003333333333333333, | |
| "loss": 0.1136, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "grad_norm": 0.3793877959251404, | |
| "learning_rate": 0.0032539682539682543, | |
| "loss": 0.2048, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "grad_norm": 0.6903932094573975, | |
| "learning_rate": 0.0031746031746031746, | |
| "loss": 0.2696, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "grad_norm": 0.628604531288147, | |
| "learning_rate": 0.0030952380952380953, | |
| "loss": 0.1585, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "grad_norm": 0.7250388860702515, | |
| "learning_rate": 0.0030158730158730157, | |
| "loss": 0.1465, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9725321888412017, | |
| "eval_f1": 0.9643166172765768, | |
| "eval_loss": 0.07334637641906738, | |
| "eval_precision": 0.9653162002256106, | |
| "eval_recall": 0.9633286722904824, | |
| "eval_runtime": 11.4298, | |
| "eval_samples_per_second": 101.927, | |
| "eval_steps_per_second": 6.387, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 0.4553186893463135, | |
| "learning_rate": 0.002936507936507937, | |
| "loss": 0.1609, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "grad_norm": 0.3900231420993805, | |
| "learning_rate": 0.002857142857142857, | |
| "loss": 0.1517, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "grad_norm": 0.45594316720962524, | |
| "learning_rate": 0.002777777777777778, | |
| "loss": 0.1705, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "grad_norm": 0.46610358357429504, | |
| "learning_rate": 0.002698412698412698, | |
| "loss": 0.1451, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "grad_norm": 0.557108461856842, | |
| "learning_rate": 0.0026190476190476194, | |
| "loss": 0.1582, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "grad_norm": 0.4486098885536194, | |
| "learning_rate": 0.0025396825396825397, | |
| "loss": 0.1763, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.9725321888412017, | |
| "eval_f1": 0.9639271421162228, | |
| "eval_loss": 0.07630692422389984, | |
| "eval_precision": 0.9702764832336926, | |
| "eval_recall": 0.9580085885445321, | |
| "eval_runtime": 11.1034, | |
| "eval_samples_per_second": 104.923, | |
| "eval_steps_per_second": 6.575, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "grad_norm": 0.5927975177764893, | |
| "learning_rate": 0.0024603174603174604, | |
| "loss": 0.1563, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "grad_norm": 0.5728869438171387, | |
| "learning_rate": 0.0023809523809523807, | |
| "loss": 0.1666, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "grad_norm": 0.34924712777137756, | |
| "learning_rate": 0.0023015873015873015, | |
| "loss": 0.1425, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "grad_norm": 0.487810343503952, | |
| "learning_rate": 0.0022222222222222222, | |
| "loss": 0.1157, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "grad_norm": 0.5470776557922363, | |
| "learning_rate": 0.002142857142857143, | |
| "loss": 0.1503, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "grad_norm": 0.9323746562004089, | |
| "learning_rate": 0.0020634920634920633, | |
| "loss": 0.1813, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "grad_norm": 0.4948749542236328, | |
| "learning_rate": 0.001984126984126984, | |
| "loss": 0.1627, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "eval_accuracy": 0.9570815450643777, | |
| "eval_f1": 0.9465564950602603, | |
| "eval_loss": 0.10568992793560028, | |
| "eval_precision": 0.9315479562643401, | |
| "eval_recall": 0.965643911608289, | |
| "eval_runtime": 11.3843, | |
| "eval_samples_per_second": 102.334, | |
| "eval_steps_per_second": 6.412, | |
| "step": 382 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "grad_norm": 0.5047426819801331, | |
| "learning_rate": 0.0019047619047619048, | |
| "loss": 0.1335, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "grad_norm": 0.3176202178001404, | |
| "learning_rate": 0.0018253968253968253, | |
| "loss": 0.0897, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "grad_norm": 0.42543891072273254, | |
| "learning_rate": 0.001746031746031746, | |
| "loss": 0.1356, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 6.59, | |
| "grad_norm": 0.35515210032463074, | |
| "learning_rate": 0.0016666666666666666, | |
| "loss": 0.1324, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "grad_norm": 0.900916337966919, | |
| "learning_rate": 0.0015873015873015873, | |
| "loss": 0.1435, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "grad_norm": 0.6653035283088684, | |
| "learning_rate": 0.0015079365079365078, | |
| "loss": 0.1509, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.975107296137339, | |
| "eval_f1": 0.9680345077825117, | |
| "eval_loss": 0.07007680833339691, | |
| "eval_precision": 0.9637800176993805, | |
| "eval_recall": 0.9725189498135582, | |
| "eval_runtime": 12.1771, | |
| "eval_samples_per_second": 95.672, | |
| "eval_steps_per_second": 5.995, | |
| "step": 446 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "grad_norm": 0.31296506524086, | |
| "learning_rate": 0.0014285714285714286, | |
| "loss": 0.1322, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "grad_norm": 0.6464645862579346, | |
| "learning_rate": 0.001349206349206349, | |
| "loss": 0.1457, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "grad_norm": 1.0385708808898926, | |
| "learning_rate": 0.0012698412698412698, | |
| "loss": 0.1278, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "grad_norm": 0.6095046401023865, | |
| "learning_rate": 0.0011904761904761904, | |
| "loss": 0.1168, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "grad_norm": 0.7724207639694214, | |
| "learning_rate": 0.0011111111111111111, | |
| "loss": 0.1182, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "grad_norm": 0.4225674569606781, | |
| "learning_rate": 0.0010317460317460316, | |
| "loss": 0.1325, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.47218218445777893, | |
| "learning_rate": 0.0009523809523809524, | |
| "loss": 0.1209, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9570815450643777, | |
| "eval_f1": 0.9465564950602603, | |
| "eval_loss": 0.10465892404317856, | |
| "eval_precision": 0.9315479562643401, | |
| "eval_recall": 0.965643911608289, | |
| "eval_runtime": 11.3686, | |
| "eval_samples_per_second": 102.475, | |
| "eval_steps_per_second": 6.421, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "grad_norm": 0.296578973531723, | |
| "learning_rate": 0.000873015873015873, | |
| "loss": 0.1006, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 8.31, | |
| "grad_norm": 0.9204122424125671, | |
| "learning_rate": 0.0007936507936507937, | |
| "loss": 0.1202, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "grad_norm": 0.5195232629776001, | |
| "learning_rate": 0.0007142857142857143, | |
| "loss": 0.1052, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "grad_norm": 0.29583191871643066, | |
| "learning_rate": 0.0006349206349206349, | |
| "loss": 0.1285, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "grad_norm": 0.4851154386997223, | |
| "learning_rate": 0.0005555555555555556, | |
| "loss": 0.1271, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "grad_norm": 0.47467851638793945, | |
| "learning_rate": 0.0004761904761904762, | |
| "loss": 0.0961, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.9733905579399141, | |
| "eval_f1": 0.966177521420591, | |
| "eval_loss": 0.07205679267644882, | |
| "eval_precision": 0.9577373642293522, | |
| "eval_recall": 0.975613576624488, | |
| "eval_runtime": 11.2412, | |
| "eval_samples_per_second": 103.637, | |
| "eval_steps_per_second": 6.494, | |
| "step": 573 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "grad_norm": 0.29292458295822144, | |
| "learning_rate": 0.0003968253968253968, | |
| "loss": 0.096, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "grad_norm": 0.513671338558197, | |
| "learning_rate": 0.00031746031746031746, | |
| "loss": 0.0871, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "grad_norm": 0.8143641948699951, | |
| "learning_rate": 0.0002380952380952381, | |
| "loss": 0.1297, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "grad_norm": 0.4731471836566925, | |
| "learning_rate": 0.00015873015873015873, | |
| "loss": 0.1175, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "grad_norm": 0.30293089151382446, | |
| "learning_rate": 7.936507936507937e-05, | |
| "loss": 0.1166, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "grad_norm": 0.3364422023296356, | |
| "learning_rate": 0.0, | |
| "loss": 0.1063, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "eval_accuracy": 0.9622317596566523, | |
| "eval_f1": 0.9526072485207102, | |
| "eval_loss": 0.08854742348194122, | |
| "eval_precision": 0.9398335630995462, | |
| "eval_recall": 0.9680642154165903, | |
| "eval_runtime": 11.4305, | |
| "eval_samples_per_second": 101.92, | |
| "eval_steps_per_second": 6.386, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "step": 630, | |
| "total_flos": 3.1439051980091965e+18, | |
| "train_loss": 0.1791202223490155, | |
| "train_runtime": 695.1527, | |
| "train_samples_per_second": 58.649, | |
| "train_steps_per_second": 0.906 | |
| } | |
| ], | |
| "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 | |
| } | |