| { | |
| "best_metric": 2.917006731033325, | |
| "best_model_checkpoint": "detr-r50-mist1-bg-8ah-4l/checkpoint-2415", | |
| "epoch": 25.0, | |
| "eval_steps": 500, | |
| "global_step": 2875, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.624347826086956e-06, | |
| "loss": 4.4466, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.812701463699341, | |
| "eval_runtime": 5.8935, | |
| "eval_samples_per_second": 6.787, | |
| "eval_steps_per_second": 0.848, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.224347826086957e-06, | |
| "loss": 3.85, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 3.863551378250122, | |
| "eval_runtime": 5.9222, | |
| "eval_samples_per_second": 6.754, | |
| "eval_steps_per_second": 0.844, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.824347826086957e-06, | |
| "loss": 3.8198, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 3.617934465408325, | |
| "eval_runtime": 5.9064, | |
| "eval_samples_per_second": 6.772, | |
| "eval_steps_per_second": 0.847, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 8.427826086956521e-06, | |
| "loss": 3.6799, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 3.455796003341675, | |
| "eval_runtime": 6.041, | |
| "eval_samples_per_second": 6.621, | |
| "eval_steps_per_second": 0.828, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 8.027826086956523e-06, | |
| "loss": 3.5806, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 3.2327613830566406, | |
| "eval_runtime": 5.9282, | |
| "eval_samples_per_second": 6.747, | |
| "eval_steps_per_second": 0.843, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 7.631304347826087e-06, | |
| "loss": 3.4958, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 3.340712070465088, | |
| "eval_runtime": 5.9472, | |
| "eval_samples_per_second": 6.726, | |
| "eval_steps_per_second": 0.841, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 7.231304347826088e-06, | |
| "loss": 3.4662, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 3.156733989715576, | |
| "eval_runtime": 5.8659, | |
| "eval_samples_per_second": 6.819, | |
| "eval_steps_per_second": 0.852, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 6.8313043478260874e-06, | |
| "loss": 3.4295, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 3.049887180328369, | |
| "eval_runtime": 5.8059, | |
| "eval_samples_per_second": 6.89, | |
| "eval_steps_per_second": 0.861, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 6.431304347826087e-06, | |
| "loss": 3.3977, | |
| "step": 1035 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 3.0459601879119873, | |
| "eval_runtime": 5.876, | |
| "eval_samples_per_second": 6.807, | |
| "eval_steps_per_second": 0.851, | |
| "step": 1035 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 6.031304347826088e-06, | |
| "loss": 3.3853, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 3.048147678375244, | |
| "eval_runtime": 5.737, | |
| "eval_samples_per_second": 6.972, | |
| "eval_steps_per_second": 0.872, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 5.631304347826088e-06, | |
| "loss": 3.3608, | |
| "step": 1265 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 3.033698558807373, | |
| "eval_runtime": 5.8795, | |
| "eval_samples_per_second": 6.803, | |
| "eval_steps_per_second": 0.85, | |
| "step": 1265 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 5.2313043478260875e-06, | |
| "loss": 3.2873, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 3.053478479385376, | |
| "eval_runtime": 5.7373, | |
| "eval_samples_per_second": 6.972, | |
| "eval_steps_per_second": 0.871, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 4.831304347826087e-06, | |
| "loss": 3.3164, | |
| "step": 1495 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 3.0139570236206055, | |
| "eval_runtime": 5.7591, | |
| "eval_samples_per_second": 6.945, | |
| "eval_steps_per_second": 0.868, | |
| "step": 1495 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 4.4313043478260875e-06, | |
| "loss": 3.2745, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 3.0666890144348145, | |
| "eval_runtime": 5.6397, | |
| "eval_samples_per_second": 7.093, | |
| "eval_steps_per_second": 0.887, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 4.031304347826087e-06, | |
| "loss": 3.2691, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 3.013432264328003, | |
| "eval_runtime": 5.7704, | |
| "eval_samples_per_second": 6.932, | |
| "eval_steps_per_second": 0.866, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 3.631304347826087e-06, | |
| "loss": 3.2735, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 3.020738124847412, | |
| "eval_runtime": 5.8757, | |
| "eval_samples_per_second": 6.808, | |
| "eval_steps_per_second": 0.851, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 3.2313043478260876e-06, | |
| "loss": 3.2718, | |
| "step": 1955 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 3.0004284381866455, | |
| "eval_runtime": 5.6729, | |
| "eval_samples_per_second": 7.051, | |
| "eval_steps_per_second": 0.881, | |
| "step": 1955 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 2.831304347826087e-06, | |
| "loss": 3.2504, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 3.1082119941711426, | |
| "eval_runtime": 5.6759, | |
| "eval_samples_per_second": 7.047, | |
| "eval_steps_per_second": 0.881, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 2.431304347826087e-06, | |
| "loss": 3.243, | |
| "step": 2185 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 2.9369025230407715, | |
| "eval_runtime": 5.8335, | |
| "eval_samples_per_second": 6.857, | |
| "eval_steps_per_second": 0.857, | |
| "step": 2185 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 2.0313043478260872e-06, | |
| "loss": 3.1669, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 2.9595999717712402, | |
| "eval_runtime": 5.8209, | |
| "eval_samples_per_second": 6.872, | |
| "eval_steps_per_second": 0.859, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 1.6347826086956523e-06, | |
| "loss": 3.1844, | |
| "step": 2415 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 2.917006731033325, | |
| "eval_runtime": 5.8161, | |
| "eval_samples_per_second": 6.877, | |
| "eval_steps_per_second": 0.86, | |
| "step": 2415 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 1.2347826086956523e-06, | |
| "loss": 3.1979, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 2.934359073638916, | |
| "eval_runtime": 5.7645, | |
| "eval_samples_per_second": 6.939, | |
| "eval_steps_per_second": 0.867, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 8.347826086956523e-07, | |
| "loss": 3.1702, | |
| "step": 2645 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 2.9261698722839355, | |
| "eval_runtime": 5.8376, | |
| "eval_samples_per_second": 6.852, | |
| "eval_steps_per_second": 0.857, | |
| "step": 2645 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 4.347826086956522e-07, | |
| "loss": 3.1738, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 2.925136089324951, | |
| "eval_runtime": 5.7559, | |
| "eval_samples_per_second": 6.949, | |
| "eval_steps_per_second": 0.869, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 3.478260869565218e-08, | |
| "loss": 3.1606, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 2.9274065494537354, | |
| "eval_runtime": 5.8001, | |
| "eval_samples_per_second": 6.896, | |
| "eval_steps_per_second": 0.862, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "step": 2875, | |
| "total_flos": 4.7279979072e+18, | |
| "train_loss": 3.4060823072350543, | |
| "train_runtime": 2285.4543, | |
| "train_samples_per_second": 5.032, | |
| "train_steps_per_second": 1.258 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2875, | |
| "num_train_epochs": 25, | |
| "save_steps": 500, | |
| "total_flos": 4.7279979072e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |