| { | |
| "best_metric": 0.33629149198532104, | |
| "best_model_checkpoint": "vit-base-food-items-v1/checkpoint-400", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 608, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06578947368421052, | |
| "grad_norm": 2.054168224334717, | |
| "learning_rate": 0.00019671052631578949, | |
| "loss": 2.2227, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13157894736842105, | |
| "grad_norm": 2.418569326400757, | |
| "learning_rate": 0.00019342105263157894, | |
| "loss": 1.7988, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.19736842105263158, | |
| "grad_norm": 2.0799572467803955, | |
| "learning_rate": 0.00019013157894736844, | |
| "loss": 1.3952, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2631578947368421, | |
| "grad_norm": 2.5012855529785156, | |
| "learning_rate": 0.00018684210526315792, | |
| "loss": 1.0071, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.32894736842105265, | |
| "grad_norm": 1.610549807548523, | |
| "learning_rate": 0.00018355263157894736, | |
| "loss": 0.8514, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.39473684210526316, | |
| "grad_norm": 2.7514488697052, | |
| "learning_rate": 0.00018026315789473684, | |
| "loss": 0.6752, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.4605263157894737, | |
| "grad_norm": 5.107870101928711, | |
| "learning_rate": 0.00017697368421052632, | |
| "loss": 0.617, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5263157894736842, | |
| "grad_norm": 1.621307611465454, | |
| "learning_rate": 0.0001736842105263158, | |
| "loss": 0.4383, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5921052631578947, | |
| "grad_norm": 2.050955057144165, | |
| "learning_rate": 0.00017039473684210527, | |
| "loss": 0.4703, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6578947368421053, | |
| "grad_norm": 3.5689868927001953, | |
| "learning_rate": 0.00016710526315789475, | |
| "loss": 0.4195, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6578947368421053, | |
| "eval_accuracy": 0.9054545454545454, | |
| "eval_loss": 0.5027927756309509, | |
| "eval_runtime": 6.6566, | |
| "eval_samples_per_second": 82.625, | |
| "eval_steps_per_second": 10.366, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7236842105263158, | |
| "grad_norm": 2.683819055557251, | |
| "learning_rate": 0.00016381578947368422, | |
| "loss": 0.3666, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.7894736842105263, | |
| "grad_norm": 2.7733426094055176, | |
| "learning_rate": 0.0001605263157894737, | |
| "loss": 0.3876, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8552631578947368, | |
| "grad_norm": 3.341937303543091, | |
| "learning_rate": 0.00015723684210526318, | |
| "loss": 0.3778, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.9210526315789473, | |
| "grad_norm": 1.0890475511550903, | |
| "learning_rate": 0.00015394736842105265, | |
| "loss": 0.3368, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.9868421052631579, | |
| "grad_norm": 3.217635154724121, | |
| "learning_rate": 0.0001506578947368421, | |
| "loss": 0.2434, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0526315789473684, | |
| "grad_norm": 1.1362298727035522, | |
| "learning_rate": 0.00014736842105263158, | |
| "loss": 0.1537, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.118421052631579, | |
| "grad_norm": 0.3043310344219208, | |
| "learning_rate": 0.00014407894736842106, | |
| "loss": 0.1786, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.1842105263157894, | |
| "grad_norm": 0.36744824051856995, | |
| "learning_rate": 0.00014078947368421053, | |
| "loss": 0.1534, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.4088458716869354, | |
| "learning_rate": 0.0001375, | |
| "loss": 0.1273, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.3157894736842106, | |
| "grad_norm": 0.22641144692897797, | |
| "learning_rate": 0.00013421052631578948, | |
| "loss": 0.1072, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.3157894736842106, | |
| "eval_accuracy": 0.8945454545454545, | |
| "eval_loss": 0.37944725155830383, | |
| "eval_runtime": 6.6833, | |
| "eval_samples_per_second": 82.295, | |
| "eval_steps_per_second": 10.324, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.381578947368421, | |
| "grad_norm": 0.14886893332004547, | |
| "learning_rate": 0.00013092105263157893, | |
| "loss": 0.0846, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.4473684210526316, | |
| "grad_norm": 0.17389647662639618, | |
| "learning_rate": 0.00012763157894736844, | |
| "loss": 0.0789, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.513157894736842, | |
| "grad_norm": 0.12492559105157852, | |
| "learning_rate": 0.00012434210526315791, | |
| "loss": 0.0605, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.5789473684210527, | |
| "grad_norm": 0.14732375741004944, | |
| "learning_rate": 0.00012105263157894738, | |
| "loss": 0.0867, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.6447368421052633, | |
| "grad_norm": 0.1113506406545639, | |
| "learning_rate": 0.00011776315789473684, | |
| "loss": 0.0436, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.7105263157894737, | |
| "grad_norm": 0.09813081473112106, | |
| "learning_rate": 0.00011447368421052632, | |
| "loss": 0.0416, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.776315789473684, | |
| "grad_norm": 6.826725006103516, | |
| "learning_rate": 0.0001111842105263158, | |
| "loss": 0.0514, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.8421052631578947, | |
| "grad_norm": 0.10619573295116425, | |
| "learning_rate": 0.00010789473684210527, | |
| "loss": 0.0601, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.9078947368421053, | |
| "grad_norm": 0.13959018886089325, | |
| "learning_rate": 0.00010460526315789475, | |
| "loss": 0.0454, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.973684210526316, | |
| "grad_norm": 0.08468258380889893, | |
| "learning_rate": 0.00010131578947368421, | |
| "loss": 0.0326, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.973684210526316, | |
| "eval_accuracy": 0.9054545454545454, | |
| "eval_loss": 0.38323774933815, | |
| "eval_runtime": 6.0691, | |
| "eval_samples_per_second": 90.622, | |
| "eval_steps_per_second": 11.369, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.039473684210526, | |
| "grad_norm": 0.07823757082223892, | |
| "learning_rate": 9.802631578947369e-05, | |
| "loss": 0.0392, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "grad_norm": 0.07656868547201157, | |
| "learning_rate": 9.473684210526316e-05, | |
| "loss": 0.0288, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.1710526315789473, | |
| "grad_norm": 0.07013211399316788, | |
| "learning_rate": 9.144736842105264e-05, | |
| "loss": 0.0313, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.236842105263158, | |
| "grad_norm": 0.07913695275783539, | |
| "learning_rate": 8.81578947368421e-05, | |
| "loss": 0.0378, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.3026315789473686, | |
| "grad_norm": 0.3869466483592987, | |
| "learning_rate": 8.486842105263159e-05, | |
| "loss": 0.0253, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.3684210526315788, | |
| "grad_norm": 0.06490592658519745, | |
| "learning_rate": 8.157894736842105e-05, | |
| "loss": 0.0241, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.4342105263157894, | |
| "grad_norm": 0.06631086021661758, | |
| "learning_rate": 7.828947368421053e-05, | |
| "loss": 0.0231, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.05489266291260719, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0218, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.5657894736842106, | |
| "grad_norm": 0.07426982372999191, | |
| "learning_rate": 7.171052631578947e-05, | |
| "loss": 0.0215, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.6315789473684212, | |
| "grad_norm": 0.063384510576725, | |
| "learning_rate": 6.842105263157895e-05, | |
| "loss": 0.0207, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.6315789473684212, | |
| "eval_accuracy": 0.9236363636363636, | |
| "eval_loss": 0.33629149198532104, | |
| "eval_runtime": 6.0608, | |
| "eval_samples_per_second": 90.746, | |
| "eval_steps_per_second": 11.385, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.6973684210526314, | |
| "grad_norm": 0.05782260745763779, | |
| "learning_rate": 6.513157894736842e-05, | |
| "loss": 0.0201, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.763157894736842, | |
| "grad_norm": 0.05535552278161049, | |
| "learning_rate": 6.18421052631579e-05, | |
| "loss": 0.0194, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.8289473684210527, | |
| "grad_norm": 0.05756945163011551, | |
| "learning_rate": 5.855263157894737e-05, | |
| "loss": 0.0191, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.8947368421052633, | |
| "grad_norm": 0.05671467259526253, | |
| "learning_rate": 5.526315789473685e-05, | |
| "loss": 0.0188, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.9605263157894735, | |
| "grad_norm": 0.05619660019874573, | |
| "learning_rate": 5.197368421052632e-05, | |
| "loss": 0.0183, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.026315789473684, | |
| "grad_norm": 0.05277419090270996, | |
| "learning_rate": 4.868421052631579e-05, | |
| "loss": 0.0177, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.0921052631578947, | |
| "grad_norm": 0.05281645059585571, | |
| "learning_rate": 4.539473684210527e-05, | |
| "loss": 0.0174, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.1578947368421053, | |
| "grad_norm": 0.06867770105600357, | |
| "learning_rate": 4.210526315789474e-05, | |
| "loss": 0.017, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.223684210526316, | |
| "grad_norm": 0.047292064875364304, | |
| "learning_rate": 3.8815789473684214e-05, | |
| "loss": 0.0168, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.2894736842105265, | |
| "grad_norm": 0.043311525136232376, | |
| "learning_rate": 3.5526315789473684e-05, | |
| "loss": 0.0167, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.2894736842105265, | |
| "eval_accuracy": 0.9236363636363636, | |
| "eval_loss": 0.33733832836151123, | |
| "eval_runtime": 5.7257, | |
| "eval_samples_per_second": 96.057, | |
| "eval_steps_per_second": 12.051, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.3552631578947367, | |
| "grad_norm": 0.04796218127012253, | |
| "learning_rate": 3.223684210526316e-05, | |
| "loss": 0.0165, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.4210526315789473, | |
| "grad_norm": 0.048424966633319855, | |
| "learning_rate": 2.8947368421052634e-05, | |
| "loss": 0.0163, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.486842105263158, | |
| "grad_norm": 0.046178512275218964, | |
| "learning_rate": 2.565789473684211e-05, | |
| "loss": 0.0157, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.5526315789473686, | |
| "grad_norm": 0.04182315245270729, | |
| "learning_rate": 2.236842105263158e-05, | |
| "loss": 0.0156, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.6184210526315788, | |
| "grad_norm": 0.04811399057507515, | |
| "learning_rate": 1.9078947368421056e-05, | |
| "loss": 0.0157, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.6842105263157894, | |
| "grad_norm": 0.04523231461644173, | |
| "learning_rate": 1.5789473684210526e-05, | |
| "loss": 0.0157, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.04799880087375641, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0155, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.8157894736842106, | |
| "grad_norm": 0.04668057709932327, | |
| "learning_rate": 9.210526315789474e-06, | |
| "loss": 0.0154, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.8815789473684212, | |
| "grad_norm": 0.044472016394138336, | |
| "learning_rate": 5.921052631578948e-06, | |
| "loss": 0.0154, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.9473684210526314, | |
| "grad_norm": 0.05030672252178192, | |
| "learning_rate": 2.631578947368421e-06, | |
| "loss": 0.0153, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.9473684210526314, | |
| "eval_accuracy": 0.9236363636363636, | |
| "eval_loss": 0.33738574385643005, | |
| "eval_runtime": 6.0053, | |
| "eval_samples_per_second": 91.586, | |
| "eval_steps_per_second": 11.49, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 608, | |
| "total_flos": 7.501829674622976e+17, | |
| "train_loss": 0.22265003621578217, | |
| "train_runtime": 237.6059, | |
| "train_samples_per_second": 40.74, | |
| "train_steps_per_second": 2.559 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 608, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.501829674622976e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |