| { | |
| "best_metric": 0.015090291388332844, | |
| "best_model_checkpoint": "./convnext-tiny-mask-finetuned/checkpoint-500", | |
| "epoch": 3.5398230088495577, | |
| "global_step": 800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00019778761061946905, | |
| "loss": 0.0126, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00019557522123893806, | |
| "loss": 0.0237, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0001933628318584071, | |
| "loss": 0.0026, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00019115044247787613, | |
| "loss": 0.0213, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00018893805309734514, | |
| "loss": 0.01, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00018672566371681417, | |
| "loss": 0.0325, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0001845132743362832, | |
| "loss": 0.0212, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00018230088495575222, | |
| "loss": 0.0138, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00018008849557522126, | |
| "loss": 0.0175, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0001778761061946903, | |
| "loss": 0.013, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_accuracy": 0.9944382647385984, | |
| "eval_loss": 0.023499028757214546, | |
| "eval_runtime": 80.0912, | |
| "eval_samples_per_second": 44.899, | |
| "eval_steps_per_second": 5.619, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0001756637168141593, | |
| "loss": 0.0188, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00017345132743362834, | |
| "loss": 0.0157, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00017123893805309737, | |
| "loss": 0.0108, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00016902654867256638, | |
| "loss": 0.0125, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00016681415929203542, | |
| "loss": 0.0165, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00016460176991150443, | |
| "loss": 0.0226, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00016238938053097346, | |
| "loss": 0.0083, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0001601769911504425, | |
| "loss": 0.0231, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0001579646017699115, | |
| "loss": 0.0151, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.00015575221238938055, | |
| "loss": 0.014, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_accuracy": 0.9936040044493882, | |
| "eval_loss": 0.01512084435671568, | |
| "eval_runtime": 80.2971, | |
| "eval_samples_per_second": 44.784, | |
| "eval_steps_per_second": 5.604, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.00015353982300884958, | |
| "loss": 0.0195, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.0001513274336283186, | |
| "loss": 0.0328, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00014911504424778763, | |
| "loss": 0.0053, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.00014690265486725664, | |
| "loss": 0.004, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.00014469026548672567, | |
| "loss": 0.021, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.00014247787610619468, | |
| "loss": 0.0078, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.00014026548672566372, | |
| "loss": 0.0149, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.00013805309734513276, | |
| "loss": 0.0149, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00013584070796460177, | |
| "loss": 0.0052, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0001336283185840708, | |
| "loss": 0.0022, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_accuracy": 0.9947163515016685, | |
| "eval_loss": 0.017131101340055466, | |
| "eval_runtime": 80.9561, | |
| "eval_samples_per_second": 44.419, | |
| "eval_steps_per_second": 5.559, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.0001314159292035398, | |
| "loss": 0.0056, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.00012920353982300885, | |
| "loss": 0.0176, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.00012699115044247788, | |
| "loss": 0.0066, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.0001247787610619469, | |
| "loss": 0.0139, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 0.00012256637168141593, | |
| "loss": 0.0056, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.00012035398230088497, | |
| "loss": 0.0009, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.00011814159292035399, | |
| "loss": 0.0004, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.00011592920353982301, | |
| "loss": 0.0094, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.00011371681415929205, | |
| "loss": 0.0037, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.00011150442477876106, | |
| "loss": 0.0022, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_accuracy": 0.9958286985539488, | |
| "eval_loss": 0.01814076118171215, | |
| "eval_runtime": 80.2281, | |
| "eval_samples_per_second": 44.822, | |
| "eval_steps_per_second": 5.609, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.00010929203539823009, | |
| "loss": 0.01, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 0.00010707964601769913, | |
| "loss": 0.0117, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.00010486725663716814, | |
| "loss": 0.0055, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.00010265486725663717, | |
| "loss": 0.0069, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00010044247787610618, | |
| "loss": 0.0102, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.823008849557522e-05, | |
| "loss": 0.0018, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 9.601769911504426e-05, | |
| "loss": 0.0027, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 9.380530973451328e-05, | |
| "loss": 0.0115, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 9.15929203539823e-05, | |
| "loss": 0.0098, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 8.938053097345133e-05, | |
| "loss": 0.0037, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "eval_accuracy": 0.9961067853170189, | |
| "eval_loss": 0.015090291388332844, | |
| "eval_runtime": 80.1059, | |
| "eval_samples_per_second": 44.891, | |
| "eval_steps_per_second": 5.618, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 8.716814159292036e-05, | |
| "loss": 0.0031, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 8.495575221238938e-05, | |
| "loss": 0.0033, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 8.274336283185841e-05, | |
| "loss": 0.0005, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 8.053097345132744e-05, | |
| "loss": 0.0008, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 7.831858407079647e-05, | |
| "loss": 0.0008, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 7.610619469026549e-05, | |
| "loss": 0.0009, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 7.389380530973453e-05, | |
| "loss": 0.0001, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 7.168141592920355e-05, | |
| "loss": 0.0004, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 6.946902654867257e-05, | |
| "loss": 0.0084, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 6.725663716814161e-05, | |
| "loss": 0.0102, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "eval_accuracy": 0.9949944382647385, | |
| "eval_loss": 0.016862528398633003, | |
| "eval_runtime": 79.8413, | |
| "eval_samples_per_second": 45.039, | |
| "eval_steps_per_second": 5.636, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 6.504424778761063e-05, | |
| "loss": 0.0002, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 6.283185840707965e-05, | |
| "loss": 0.0035, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 6.061946902654868e-05, | |
| "loss": 0.0017, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 5.8407079646017705e-05, | |
| "loss": 0.0004, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 5.619469026548673e-05, | |
| "loss": 0.0001, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 5.398230088495575e-05, | |
| "loss": 0.0, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 5.176991150442479e-05, | |
| "loss": 0.0003, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 4.955752212389381e-05, | |
| "loss": 0.0005, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 4.734513274336283e-05, | |
| "loss": 0.0001, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 4.5132743362831855e-05, | |
| "loss": 0.0003, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "eval_accuracy": 0.9961067853170189, | |
| "eval_loss": 0.01964888535439968, | |
| "eval_runtime": 79.8617, | |
| "eval_samples_per_second": 45.028, | |
| "eval_steps_per_second": 5.635, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 4.2920353982300885e-05, | |
| "loss": 0.0001, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 4.0707964601769914e-05, | |
| "loss": 0.0007, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 3.849557522123894e-05, | |
| "loss": 0.0001, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 3.628318584070797e-05, | |
| "loss": 0.0017, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 3.407079646017699e-05, | |
| "loss": 0.0006, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 3.185840707964602e-05, | |
| "loss": 0.0003, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 2.964601769911505e-05, | |
| "loss": 0.0001, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 2.743362831858407e-05, | |
| "loss": 0.0001, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 2.5221238938053098e-05, | |
| "loss": 0.0008, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 2.3008849557522124e-05, | |
| "loss": 0.0002, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "eval_accuracy": 0.9955506117908788, | |
| "eval_loss": 0.017847377806901932, | |
| "eval_runtime": 81.2034, | |
| "eval_samples_per_second": 44.284, | |
| "eval_steps_per_second": 5.542, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "step": 800, | |
| "total_flos": 1.2838213030685737e+18, | |
| "train_loss": 0.007916164267808199, | |
| "train_runtime": 4100.2539, | |
| "train_samples_per_second": 14.075, | |
| "train_steps_per_second": 0.22 | |
| } | |
| ], | |
| "max_steps": 904, | |
| "num_train_epochs": 4, | |
| "total_flos": 1.2838213030685737e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |