| { | |
| "best_metric": 0.8969686033922771, | |
| "best_model_checkpoint": "violation-classification-bantai-vit-v100ep/checkpoint-808", | |
| "epoch": 8.99753086419753, | |
| "global_step": 909, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.950495049504951e-07, | |
| "loss": 1.3728, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.900990099009902e-07, | |
| "loss": 1.37, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 1.4851485148514852e-06, | |
| "loss": 1.3698, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 1.9801980198019803e-06, | |
| "loss": 1.3582, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 2.4752475247524753e-06, | |
| "loss": 1.348, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 2.9702970297029703e-06, | |
| "loss": 1.3355, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 3.4653465346534657e-06, | |
| "loss": 1.3186, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 3.960396039603961e-06, | |
| "loss": 1.3057, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.455445544554456e-06, | |
| "loss": 1.2756, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.950495049504951e-06, | |
| "loss": 1.2596, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5615301335258029, | |
| "eval_loss": 1.222981333732605, | |
| "eval_runtime": 59.3749, | |
| "eval_samples_per_second": 93.339, | |
| "eval_steps_per_second": 2.931, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 5.445544554455446e-06, | |
| "loss": 1.2563, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 5.940594059405941e-06, | |
| "loss": 1.1843, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 6.4356435643564364e-06, | |
| "loss": 1.1572, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 6.9306930693069314e-06, | |
| "loss": 1.1153, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 7.4257425742574256e-06, | |
| "loss": 1.0653, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 7.920792079207921e-06, | |
| "loss": 1.0119, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 8.415841584158417e-06, | |
| "loss": 0.9734, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 8.910891089108911e-06, | |
| "loss": 0.9144, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 9.405940594059407e-06, | |
| "loss": 0.9037, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 9.900990099009901e-06, | |
| "loss": 0.8527, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6840490797546013, | |
| "eval_loss": 0.8233989477157593, | |
| "eval_runtime": 59.669, | |
| "eval_samples_per_second": 92.879, | |
| "eval_steps_per_second": 2.916, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.0396039603960395e-05, | |
| "loss": 0.8402, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 1.0891089108910891e-05, | |
| "loss": 0.7847, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 1.1386138613861387e-05, | |
| "loss": 0.7738, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 1.1881188118811881e-05, | |
| "loss": 0.7346, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 1.2376237623762377e-05, | |
| "loss": 0.6992, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 1.2871287128712873e-05, | |
| "loss": 0.695, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 1.3366336633663367e-05, | |
| "loss": 0.6715, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 1.3861386138613863e-05, | |
| "loss": 0.6683, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 1.4356435643564355e-05, | |
| "loss": 0.6617, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 1.4851485148514851e-05, | |
| "loss": 0.6375, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.784554312522555, | |
| "eval_loss": 0.6000781655311584, | |
| "eval_runtime": 59.2996, | |
| "eval_samples_per_second": 93.458, | |
| "eval_steps_per_second": 2.934, | |
| "step": 303 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 1.534653465346535e-05, | |
| "loss": 0.6349, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 1.5841584158415843e-05, | |
| "loss": 0.5924, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 1.6336633663366337e-05, | |
| "loss": 0.5905, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 1.6831683168316834e-05, | |
| "loss": 0.5591, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 1.7326732673267325e-05, | |
| "loss": 0.582, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 1.7821782178217823e-05, | |
| "loss": 0.553, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 1.8316831683168317e-05, | |
| "loss": 0.5489, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 1.8811881188118814e-05, | |
| "loss": 0.5711, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 1.930693069306931e-05, | |
| "loss": 0.5161, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 1.9801980198019803e-05, | |
| "loss": 0.555, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.817755322988091, | |
| "eval_loss": 0.5038326978683472, | |
| "eval_runtime": 59.2865, | |
| "eval_samples_per_second": 93.478, | |
| "eval_steps_per_second": 2.935, | |
| "step": 404 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 2.02970297029703e-05, | |
| "loss": 0.5553, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 2.079207920792079e-05, | |
| "loss": 0.5075, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 2.128712871287129e-05, | |
| "loss": 0.4948, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 2.1782178217821783e-05, | |
| "loss": 0.4892, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 2.227722772277228e-05, | |
| "loss": 0.4861, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 2.2772277227722774e-05, | |
| "loss": 0.4922, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 2.326732673267327e-05, | |
| "loss": 0.498, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 2.3762376237623762e-05, | |
| "loss": 0.4722, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "learning_rate": 2.4257425742574257e-05, | |
| "loss": 0.4619, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 2.4752475247524754e-05, | |
| "loss": 0.4433, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.843558282208589, | |
| "eval_loss": 0.4338299632072449, | |
| "eval_runtime": 59.1883, | |
| "eval_samples_per_second": 93.633, | |
| "eval_steps_per_second": 2.94, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 2.5247524752475248e-05, | |
| "loss": 0.4457, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 2.5742574257425746e-05, | |
| "loss": 0.4402, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "learning_rate": 2.623762376237624e-05, | |
| "loss": 0.4118, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 2.6732673267326734e-05, | |
| "loss": 0.4431, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 2.722772277227723e-05, | |
| "loss": 0.4308, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 2.7722772277227726e-05, | |
| "loss": 0.4262, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "learning_rate": 2.8217821782178216e-05, | |
| "loss": 0.4131, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "learning_rate": 2.871287128712871e-05, | |
| "loss": 0.3946, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 2.9207920792079208e-05, | |
| "loss": 0.4015, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "learning_rate": 2.9702970297029702e-05, | |
| "loss": 0.406, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8661133164922411, | |
| "eval_loss": 0.3765123784542084, | |
| "eval_runtime": 59.3663, | |
| "eval_samples_per_second": 93.353, | |
| "eval_steps_per_second": 2.931, | |
| "step": 606 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 3.01980198019802e-05, | |
| "loss": 0.4215, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 3.06930693069307e-05, | |
| "loss": 0.3587, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 3.118811881188119e-05, | |
| "loss": 0.3565, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 3.1683168316831686e-05, | |
| "loss": 0.3848, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 3.217821782178218e-05, | |
| "loss": 0.3508, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 3.2673267326732674e-05, | |
| "loss": 0.3723, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 3.3168316831683175e-05, | |
| "loss": 0.3643, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 3.366336633663367e-05, | |
| "loss": 0.4062, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 3.415841584158416e-05, | |
| "loss": 0.3767, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 3.465346534653465e-05, | |
| "loss": 0.3517, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8792854565138939, | |
| "eval_loss": 0.34659793972969055, | |
| "eval_runtime": 59.2647, | |
| "eval_samples_per_second": 93.513, | |
| "eval_steps_per_second": 2.936, | |
| "step": 707 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 3.514851485148515e-05, | |
| "loss": 0.3314, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "learning_rate": 3.5643564356435645e-05, | |
| "loss": 0.3319, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 3.613861386138614e-05, | |
| "loss": 0.32, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 3.6633663366336634e-05, | |
| "loss": 0.3448, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 3.712871287128713e-05, | |
| "loss": 0.3242, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 3.762376237623763e-05, | |
| "loss": 0.3241, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 3.811881188118812e-05, | |
| "loss": 0.3379, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 3.861386138613862e-05, | |
| "loss": 0.3392, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 7.82, | |
| "learning_rate": 3.910891089108911e-05, | |
| "loss": 0.2918, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 7.92, | |
| "learning_rate": 3.9603960396039605e-05, | |
| "loss": 0.312, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8969686033922771, | |
| "eval_loss": 0.3010924756526947, | |
| "eval_runtime": 59.78, | |
| "eval_samples_per_second": 92.707, | |
| "eval_steps_per_second": 2.911, | |
| "step": 808 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 4.0099009900990106e-05, | |
| "loss": 0.2918, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 4.05940594059406e-05, | |
| "loss": 0.308, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 4.108910891089109e-05, | |
| "loss": 0.2897, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "learning_rate": 4.158415841584158e-05, | |
| "loss": 0.2688, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 4.207920792079208e-05, | |
| "loss": 0.2767, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 4.257425742574258e-05, | |
| "loss": 0.2737, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 4.306930693069307e-05, | |
| "loss": 0.2805, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 4.3564356435643565e-05, | |
| "loss": 0.2799, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 8.81, | |
| "learning_rate": 4.405940594059406e-05, | |
| "loss": 0.2634, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 8.91, | |
| "learning_rate": 4.455445544554456e-05, | |
| "loss": 0.2842, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.896066402020931, | |
| "eval_loss": 0.2943052351474762, | |
| "eval_runtime": 59.263, | |
| "eval_samples_per_second": 93.515, | |
| "eval_steps_per_second": 2.936, | |
| "step": 909 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "step": 909, | |
| "total_flos": 9.017904477803397e+18, | |
| "train_loss": 0.6144383929350196, | |
| "train_runtime": 3112.0693, | |
| "train_samples_per_second": 415.479, | |
| "train_steps_per_second": 3.245 | |
| } | |
| ], | |
| "max_steps": 10100, | |
| "num_train_epochs": 100, | |
| "total_flos": 9.017904477803397e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |