| { | |
| "best_metric": 0.9053803339517625, | |
| "best_model_checkpoint": "resnet-50-resnet50_fashion/checkpoint-692", | |
| "epoch": 9.933993399339935, | |
| "global_step": 860, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.994124559341951e-05, | |
| "loss": 0.6959, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.935370152761457e-05, | |
| "loss": 0.6903, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.876615746180964e-05, | |
| "loss": 0.6815, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.81786133960047e-05, | |
| "loss": 0.6753, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.759106933019977e-05, | |
| "loss": 0.6688, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.700352526439483e-05, | |
| "loss": 0.6613, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.6415981198589895e-05, | |
| "loss": 0.6599, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.582843713278496e-05, | |
| "loss": 0.6396, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.7346938775510204, | |
| "eval_loss": 0.7625104188919067, | |
| "eval_runtime": 58.978, | |
| "eval_samples_per_second": 9.139, | |
| "eval_steps_per_second": 1.153, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 4.524089306698003e-05, | |
| "loss": 0.6247, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.465334900117509e-05, | |
| "loss": 0.628, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 4.4065804935370154e-05, | |
| "loss": 0.632, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 4.347826086956522e-05, | |
| "loss": 0.6004, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 4.289071680376029e-05, | |
| "loss": 0.5901, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 4.2303172737955346e-05, | |
| "loss": 0.5947, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 4.171562867215041e-05, | |
| "loss": 0.5753, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 4.112808460634548e-05, | |
| "loss": 0.5851, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 4.0540540540540545e-05, | |
| "loss": 0.5646, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8348794063079777, | |
| "eval_loss": 0.5780991911888123, | |
| "eval_runtime": 9.3608, | |
| "eval_samples_per_second": 57.581, | |
| "eval_steps_per_second": 7.264, | |
| "step": 173 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 3.9952996474735605e-05, | |
| "loss": 0.547, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 3.936545240893067e-05, | |
| "loss": 0.5548, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 3.877790834312574e-05, | |
| "loss": 0.5498, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 3.81903642773208e-05, | |
| "loss": 0.5291, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 3.760282021151586e-05, | |
| "loss": 0.5145, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 3.701527614571093e-05, | |
| "loss": 0.4981, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 3.6427732079905996e-05, | |
| "loss": 0.5022, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 3.584018801410106e-05, | |
| "loss": 0.4768, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.47913119196891785, | |
| "eval_runtime": 8.4707, | |
| "eval_samples_per_second": 63.631, | |
| "eval_steps_per_second": 8.028, | |
| "step": 259 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.525264394829612e-05, | |
| "loss": 0.4758, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 3.466509988249119e-05, | |
| "loss": 0.4699, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 3.4077555816686255e-05, | |
| "loss": 0.4638, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 3.3490011750881314e-05, | |
| "loss": 0.4378, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 3.290246768507638e-05, | |
| "loss": 0.4795, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 3.231492361927145e-05, | |
| "loss": 0.4522, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 3.172737955346651e-05, | |
| "loss": 0.4456, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 3.113983548766158e-05, | |
| "loss": 0.4489, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 3.055229142185664e-05, | |
| "loss": 0.4161, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8905380333951762, | |
| "eval_loss": 0.38660189509391785, | |
| "eval_runtime": 8.0364, | |
| "eval_samples_per_second": 67.07, | |
| "eval_steps_per_second": 8.461, | |
| "step": 346 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 2.9964747356051702e-05, | |
| "loss": 0.415, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 2.9377203290246768e-05, | |
| "loss": 0.4333, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 2.8789659224441835e-05, | |
| "loss": 0.4304, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 2.82021151586369e-05, | |
| "loss": 0.4151, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 2.7614571092831964e-05, | |
| "loss": 0.3846, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 2.702702702702703e-05, | |
| "loss": 0.4447, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 2.6439482961222096e-05, | |
| "loss": 0.4253, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "learning_rate": 2.5851938895417156e-05, | |
| "loss": 0.3669, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 2.526439482961222e-05, | |
| "loss": 0.402, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.9035250463821892, | |
| "eval_loss": 0.3293728232383728, | |
| "eval_runtime": 7.9465, | |
| "eval_samples_per_second": 67.829, | |
| "eval_steps_per_second": 8.557, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 2.4676850763807285e-05, | |
| "loss": 0.4157, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "learning_rate": 2.408930669800235e-05, | |
| "loss": 0.372, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 2.3501762632197415e-05, | |
| "loss": 0.4115, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 2.291421856639248e-05, | |
| "loss": 0.3954, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 2.2326674500587544e-05, | |
| "loss": 0.395, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 2.173913043478261e-05, | |
| "loss": 0.378, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 2.1151586368977673e-05, | |
| "loss": 0.354, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 2.056404230317274e-05, | |
| "loss": 0.369, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8923933209647495, | |
| "eval_loss": 1.0405044555664062, | |
| "eval_runtime": 7.8895, | |
| "eval_samples_per_second": 68.318, | |
| "eval_steps_per_second": 8.619, | |
| "step": 519 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 1.9976498237367802e-05, | |
| "loss": 0.3544, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 1.938895417156287e-05, | |
| "loss": 0.3565, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 1.880141010575793e-05, | |
| "loss": 0.3868, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 1.8213866039952998e-05, | |
| "loss": 0.4005, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "learning_rate": 1.762632197414806e-05, | |
| "loss": 0.3524, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 1.7038777908343127e-05, | |
| "loss": 0.4008, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "learning_rate": 1.645123384253819e-05, | |
| "loss": 0.3571, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "learning_rate": 1.5863689776733257e-05, | |
| "loss": 0.3398, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 1.527614571092832e-05, | |
| "loss": 0.3512, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8905380333951762, | |
| "eval_loss": 1.4846545457839966, | |
| "eval_runtime": 8.0342, | |
| "eval_samples_per_second": 67.089, | |
| "eval_steps_per_second": 8.464, | |
| "step": 606 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 1.4688601645123384e-05, | |
| "loss": 0.3451, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "learning_rate": 1.410105757931845e-05, | |
| "loss": 0.3489, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 7.28, | |
| "learning_rate": 1.3513513513513515e-05, | |
| "loss": 0.3414, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "learning_rate": 1.2925969447708578e-05, | |
| "loss": 0.3867, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 1.2338425381903643e-05, | |
| "loss": 0.3509, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 1.1750881316098707e-05, | |
| "loss": 0.33, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 1.1163337250293772e-05, | |
| "loss": 0.3678, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 1.0575793184488837e-05, | |
| "loss": 0.3481, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "learning_rate": 9.988249118683901e-06, | |
| "loss": 0.3439, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "eval_accuracy": 0.9053803339517625, | |
| "eval_loss": 0.28196877241134644, | |
| "eval_runtime": 8.0187, | |
| "eval_samples_per_second": 67.218, | |
| "eval_steps_per_second": 8.48, | |
| "step": 692 | |
| }, | |
| { | |
| "epoch": 8.09, | |
| "learning_rate": 9.400705052878966e-06, | |
| "loss": 0.3763, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 8.81316098707403e-06, | |
| "loss": 0.3435, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "learning_rate": 8.225616921269095e-06, | |
| "loss": 0.3455, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 8.43, | |
| "learning_rate": 7.63807285546416e-06, | |
| "loss": 0.3422, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 7.050528789659225e-06, | |
| "loss": 0.3484, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 6.462984723854289e-06, | |
| "loss": 0.3342, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 5.875440658049354e-06, | |
| "loss": 0.3657, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 5.287896592244418e-06, | |
| "loss": 0.3306, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8849721706864564, | |
| "eval_loss": 0.3021787703037262, | |
| "eval_runtime": 8.1523, | |
| "eval_samples_per_second": 66.116, | |
| "eval_steps_per_second": 8.341, | |
| "step": 779 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 4.700352526439483e-06, | |
| "loss": 0.3541, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 9.13, | |
| "learning_rate": 4.1128084606345476e-06, | |
| "loss": 0.3233, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 9.24, | |
| "learning_rate": 3.5252643948296126e-06, | |
| "loss": 0.3345, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "learning_rate": 2.937720329024677e-06, | |
| "loss": 0.3387, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 2.3501762632197415e-06, | |
| "loss": 0.3477, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "learning_rate": 1.7626321974148063e-06, | |
| "loss": 0.3148, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 9.7, | |
| "learning_rate": 1.1750881316098707e-06, | |
| "loss": 0.337, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "learning_rate": 5.875440658049354e-07, | |
| "loss": 0.3339, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 0.0, | |
| "loss": 0.3691, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "eval_accuracy": 0.8163265306122449, | |
| "eval_loss": 0.7981617450714111, | |
| "eval_runtime": 9.506, | |
| "eval_samples_per_second": 56.701, | |
| "eval_steps_per_second": 7.153, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "step": 860, | |
| "total_flos": 1.0221236731922227e+18, | |
| "train_loss": 0.4434790275817694, | |
| "train_runtime": 1771.1031, | |
| "train_samples_per_second": 27.356, | |
| "train_steps_per_second": 0.486 | |
| } | |
| ], | |
| "max_steps": 860, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.0221236731922227e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |