| { | |
| "best_metric": 0.9793991416309012, | |
| "best_model_checkpoint": "swin-large-patch4-window7-224-in22k-finetuned-lora-medmnistv2/checkpoint-255", | |
| "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": 0.9944203495979309, | |
| "learning_rate": 0.004920634920634921, | |
| "loss": 0.6411, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 1.986366868019104, | |
| "learning_rate": 0.004841269841269842, | |
| "loss": 0.3966, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 1.2478023767471313, | |
| "learning_rate": 0.0047619047619047615, | |
| "loss": 0.2835, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.9372655749320984, | |
| "learning_rate": 0.004682539682539683, | |
| "loss": 0.2194, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.834052562713623, | |
| "learning_rate": 0.004603174603174603, | |
| "loss": 0.4268, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.8572702407836914, | |
| "learning_rate": 0.004523809523809524, | |
| "loss": 0.3305, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.936480686695279, | |
| "eval_f1": 0.9119407558733402, | |
| "eval_loss": 0.16002865135669708, | |
| "eval_precision": 0.9478312410980476, | |
| "eval_recall": 0.886801607677731, | |
| "eval_runtime": 18.303, | |
| "eval_samples_per_second": 63.651, | |
| "eval_steps_per_second": 3.988, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 1.092854619026184, | |
| "learning_rate": 0.0044444444444444444, | |
| "loss": 0.2775, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.632826328277588, | |
| "learning_rate": 0.004365079365079365, | |
| "loss": 0.2432, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 0.9421872496604919, | |
| "learning_rate": 0.004285714285714286, | |
| "loss": 0.2564, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "grad_norm": 3.025440216064453, | |
| "learning_rate": 0.004206349206349207, | |
| "loss": 0.2726, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 0.8582188487052917, | |
| "learning_rate": 0.0041269841269841265, | |
| "loss": 0.2958, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 1.093877911567688, | |
| "learning_rate": 0.004047619047619048, | |
| "loss": 0.2335, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.9313304721030042, | |
| "eval_f1": 0.9166446055107573, | |
| "eval_loss": 0.15517598390579224, | |
| "eval_precision": 0.8967545322648443, | |
| "eval_recall": 0.9471582920716426, | |
| "eval_runtime": 18.7183, | |
| "eval_samples_per_second": 62.239, | |
| "eval_steps_per_second": 3.9, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 1.588196873664856, | |
| "learning_rate": 0.003968253968253968, | |
| "loss": 0.1984, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.8592113256454468, | |
| "learning_rate": 0.003896825396825397, | |
| "loss": 0.2741, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 1.1290347576141357, | |
| "learning_rate": 0.003817460317460317, | |
| "loss": 0.2253, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "grad_norm": 0.8879281878471375, | |
| "learning_rate": 0.0037380952380952383, | |
| "loss": 0.2726, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 0.8666319847106934, | |
| "learning_rate": 0.0036587301587301586, | |
| "loss": 0.2226, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "grad_norm": 1.2298210859298706, | |
| "learning_rate": 0.0035793650793650793, | |
| "loss": 0.1967, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "grad_norm": 0.7905568480491638, | |
| "learning_rate": 0.0034999999999999996, | |
| "loss": 0.1977, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9733905579399141, | |
| "eval_f1": 0.9659005824707372, | |
| "eval_loss": 0.08546662330627441, | |
| "eval_precision": 0.9607843137254902, | |
| "eval_recall": 0.9713575096277278, | |
| "eval_runtime": 18.7005, | |
| "eval_samples_per_second": 62.298, | |
| "eval_steps_per_second": 3.904, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "grad_norm": 0.9821394681930542, | |
| "learning_rate": 0.003420634920634921, | |
| "loss": 0.2158, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "grad_norm": 0.6182402968406677, | |
| "learning_rate": 0.003341269841269841, | |
| "loss": 0.1528, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "grad_norm": 0.6537771821022034, | |
| "learning_rate": 0.003261904761904762, | |
| "loss": 0.2038, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "grad_norm": 1.0976545810699463, | |
| "learning_rate": 0.0031825396825396826, | |
| "loss": 0.217, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "grad_norm": 0.581605076789856, | |
| "learning_rate": 0.0031031746031746034, | |
| "loss": 0.2044, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "grad_norm": 0.8219888806343079, | |
| "learning_rate": 0.0030238095238095237, | |
| "loss": 0.1746, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9793991416309012, | |
| "eval_f1": 0.972945356587167, | |
| "eval_loss": 0.0870010182261467, | |
| "eval_precision": 0.9794167490467448, | |
| "eval_recall": 0.9669104162846139, | |
| "eval_runtime": 18.7359, | |
| "eval_samples_per_second": 62.18, | |
| "eval_steps_per_second": 3.896, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 0.45729899406433105, | |
| "learning_rate": 0.0029444444444444444, | |
| "loss": 0.1785, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "grad_norm": 0.9842467904090881, | |
| "learning_rate": 0.002865079365079365, | |
| "loss": 0.1507, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "grad_norm": 2.326084852218628, | |
| "learning_rate": 0.002785714285714286, | |
| "loss": 0.2188, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "grad_norm": 0.7227071523666382, | |
| "learning_rate": 0.002706349206349206, | |
| "loss": 0.1853, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "grad_norm": 1.1356016397476196, | |
| "learning_rate": 0.002626984126984127, | |
| "loss": 0.1671, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "grad_norm": 0.6211657524108887, | |
| "learning_rate": 0.0025476190476190477, | |
| "loss": 0.1797, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.9699570815450643, | |
| "eval_f1": 0.961658208434536, | |
| "eval_loss": 0.0828637108206749, | |
| "eval_precision": 0.95490488153731, | |
| "eval_recall": 0.969034629256067, | |
| "eval_runtime": 18.2407, | |
| "eval_samples_per_second": 63.868, | |
| "eval_steps_per_second": 4.002, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "grad_norm": 0.3646533787250519, | |
| "learning_rate": 0.0024682539682539684, | |
| "loss": 0.1636, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "grad_norm": 0.5121908187866211, | |
| "learning_rate": 0.002388888888888889, | |
| "loss": 0.1463, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "grad_norm": 0.34603381156921387, | |
| "learning_rate": 0.0023095238095238095, | |
| "loss": 0.1284, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "grad_norm": 1.5245040655136108, | |
| "learning_rate": 0.0022301587301587302, | |
| "loss": 0.1307, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "grad_norm": 1.0758203268051147, | |
| "learning_rate": 0.002150793650793651, | |
| "loss": 0.1912, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "grad_norm": 0.9437044858932495, | |
| "learning_rate": 0.0020714285714285717, | |
| "loss": 0.1942, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "grad_norm": 0.6630730628967285, | |
| "learning_rate": 0.001992063492063492, | |
| "loss": 0.1436, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "eval_accuracy": 0.9708154506437768, | |
| "eval_f1": 0.9627915802446869, | |
| "eval_loss": 0.0797128826379776, | |
| "eval_precision": 0.9556120562130177, | |
| "eval_recall": 0.9706793660981723, | |
| "eval_runtime": 18.3826, | |
| "eval_samples_per_second": 63.375, | |
| "eval_steps_per_second": 3.971, | |
| "step": 382 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "grad_norm": 0.5699043869972229, | |
| "learning_rate": 0.0019126984126984126, | |
| "loss": 0.1704, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "grad_norm": 0.6287882328033447, | |
| "learning_rate": 0.0018333333333333333, | |
| "loss": 0.1034, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "grad_norm": 0.4099302589893341, | |
| "learning_rate": 0.0017539682539682538, | |
| "loss": 0.1633, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 6.59, | |
| "grad_norm": 0.24668000638484955, | |
| "learning_rate": 0.0016746031746031746, | |
| "loss": 0.1477, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "grad_norm": 0.6959215998649597, | |
| "learning_rate": 0.001595238095238095, | |
| "loss": 0.1578, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "grad_norm": 0.4429934620857239, | |
| "learning_rate": 0.0015158730158730158, | |
| "loss": 0.1632, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9699570815450643, | |
| "eval_f1": 0.9621164611060785, | |
| "eval_loss": 0.08162170648574829, | |
| "eval_precision": 0.9507741239032943, | |
| "eval_recall": 0.9754187297512072, | |
| "eval_runtime": 18.5501, | |
| "eval_samples_per_second": 62.803, | |
| "eval_steps_per_second": 3.935, | |
| "step": 446 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "grad_norm": 0.3358498513698578, | |
| "learning_rate": 0.0014365079365079364, | |
| "loss": 0.1628, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "grad_norm": 1.001493215560913, | |
| "learning_rate": 0.0013571428571428571, | |
| "loss": 0.1495, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "grad_norm": 1.0969635248184204, | |
| "learning_rate": 0.0012777777777777776, | |
| "loss": 0.1972, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "grad_norm": 0.4329684376716614, | |
| "learning_rate": 0.0011984126984126984, | |
| "loss": 0.1538, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "grad_norm": 0.426087886095047, | |
| "learning_rate": 0.0011190476190476191, | |
| "loss": 0.1291, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "grad_norm": 0.24835826456546783, | |
| "learning_rate": 0.0010396825396825396, | |
| "loss": 0.1301, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.3314363956451416, | |
| "learning_rate": 0.0009603174603174604, | |
| "loss": 0.1125, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9613733905579399, | |
| "eval_f1": 0.9519463889441405, | |
| "eval_loss": 0.10071194916963577, | |
| "eval_precision": 0.9364616472251679, | |
| "eval_recall": 0.9717395623204352, | |
| "eval_runtime": 18.4952, | |
| "eval_samples_per_second": 62.989, | |
| "eval_steps_per_second": 3.947, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "grad_norm": 0.4465714693069458, | |
| "learning_rate": 0.000880952380952381, | |
| "loss": 0.1279, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 8.31, | |
| "grad_norm": 0.7344756126403809, | |
| "learning_rate": 0.0008015873015873017, | |
| "loss": 0.1408, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "grad_norm": 2.2782132625579834, | |
| "learning_rate": 0.0007222222222222222, | |
| "loss": 0.134, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "grad_norm": 0.8408999443054199, | |
| "learning_rate": 0.0006428571428571428, | |
| "loss": 0.1193, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "grad_norm": 0.4357975721359253, | |
| "learning_rate": 0.0005634920634920636, | |
| "loss": 0.114, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "grad_norm": 0.6790524125099182, | |
| "learning_rate": 0.00048412698412698415, | |
| "loss": 0.1076, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.9690987124463519, | |
| "eval_f1": 0.9612241124260354, | |
| "eval_loss": 0.09004171937704086, | |
| "eval_precision": 0.9481984892871503, | |
| "eval_recall": 0.9769660431566722, | |
| "eval_runtime": 18.3674, | |
| "eval_samples_per_second": 63.427, | |
| "eval_steps_per_second": 3.974, | |
| "step": 573 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "grad_norm": 0.38498663902282715, | |
| "learning_rate": 0.0004047619047619048, | |
| "loss": 0.1174, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "grad_norm": 0.7055544853210449, | |
| "learning_rate": 0.0003253968253968254, | |
| "loss": 0.0951, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "grad_norm": 0.8157325387001038, | |
| "learning_rate": 0.00024603174603174605, | |
| "loss": 0.1397, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "grad_norm": 0.6511685252189636, | |
| "learning_rate": 0.00016666666666666666, | |
| "loss": 0.1296, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "grad_norm": 0.5405935049057007, | |
| "learning_rate": 8.730158730158731e-05, | |
| "loss": 0.1028, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "grad_norm": 0.472501665353775, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.1188, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "eval_accuracy": 0.9622317596566523, | |
| "eval_f1": 0.9529697156530292, | |
| "eval_loss": 0.10635051876306534, | |
| "eval_precision": 0.9377354615755036, | |
| "eval_recall": 0.9723202824133504, | |
| "eval_runtime": 18.4162, | |
| "eval_samples_per_second": 63.26, | |
| "eval_steps_per_second": 3.964, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "step": 630, | |
| "total_flos": 7.138406630605308e+18, | |
| "train_loss": 0.19363035304205758, | |
| "train_runtime": 1305.2945, | |
| "train_samples_per_second": 31.234, | |
| "train_steps_per_second": 0.483 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 630, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 7.138406630605308e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |