| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "best_supernet_model_checkpoint": null, | |
| "epoch": 6.0, | |
| "global_step": 19644, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 0.15, | |
| "learning_rate": 1.9968428586788752e-05, | |
| "loss": 1.2712, | |
| "step": 500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 0.31, | |
| "learning_rate": 1.987315308068463e-05, | |
| "loss": 0.9806, | |
| "step": 1000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 0.46, | |
| "learning_rate": 1.971478120799466e-05, | |
| "loss": 0.9157, | |
| "step": 1500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 0.61, | |
| "learning_rate": 1.949432507596102e-05, | |
| "loss": 0.8603, | |
| "step": 2000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 0.76, | |
| "learning_rate": 1.921319355374732e-05, | |
| "loss": 0.8511, | |
| "step": 2500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 0.92, | |
| "learning_rate": 1.887392066743481e-05, | |
| "loss": 0.8032, | |
| "step": 3000 | |
| }, | |
| { | |
| "Minimum SubNet": "OrderedDict([(<ElasticityDim.WIDTH: 'width'>, {0: 512, 1: 320, 2: 320, 3: 576, 4: 576, 5: 512, 6: 448, 7: 448, 8: 448, 9: 320, 10: 384, 11: 448, 12: 1104, 13: 1066, 14: 1126, 15: 1102, 16: 1067, 17: 1023, 18: 1048, 19: 1061, 20: 984, 21: 772, 22: 609, 23: 205})])", | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8890719384953323, | |
| "eval_loss": 0.3191866874694824, | |
| "eval_runtime": 3.7698, | |
| "eval_samples_per_second": 1449.15, | |
| "eval_steps_per_second": 11.406, | |
| "step": 3274 | |
| }, | |
| { | |
| "SuperNet": "OrderedDict([(<ElasticityDim.WIDTH: 'width'>, {0: 768, 1: 768, 2: 768, 3: 768, 4: 768, 5: 768, 6: 768, 7: 768, 8: 768, 9: 768, 10: 768, 11: 768, 12: 3072, 13: 3072, 14: 3072, 15: 3072, 16: 3072, 17: 3072, 18: 3072, 19: 3072, 20: 3072, 21: 3072, 22: 3072, 23: 3072})])", | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.90591250228812, | |
| "eval_loss": 0.2594413161277771, | |
| "eval_runtime": 4.6342, | |
| "eval_samples_per_second": 1178.834, | |
| "eval_steps_per_second": 9.279, | |
| "step": 3274 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 1.07, | |
| "learning_rate": 1.847816370500993e-05, | |
| "loss": 0.681, | |
| "step": 3500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 1.22, | |
| "learning_rate": 1.8027488163924413e-05, | |
| "loss": 0.532, | |
| "step": 4000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 1.37, | |
| "learning_rate": 1.7525511348381253e-05, | |
| "loss": 0.5419, | |
| "step": 4500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 1.53, | |
| "learning_rate": 1.6976587091762227e-05, | |
| "loss": 0.5282, | |
| "step": 5000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 1.68, | |
| "learning_rate": 1.638202448226643e-05, | |
| "loss": 0.5355, | |
| "step": 5500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 1.83, | |
| "learning_rate": 1.574667626426217e-05, | |
| "loss": 0.5507, | |
| "step": 6000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 1.99, | |
| "learning_rate": 1.507460275806913e-05, | |
| "loss": 0.5165, | |
| "step": 6500 | |
| }, | |
| { | |
| "Minimum SubNet": "OrderedDict([(<ElasticityDim.WIDTH: 'width'>, {0: 512, 1: 320, 2: 320, 3: 576, 4: 576, 5: 512, 6: 448, 7: 448, 8: 448, 9: 320, 10: 384, 11: 448, 12: 1104, 13: 1066, 14: 1126, 15: 1102, 16: 1067, 17: 1023, 18: 1048, 19: 1061, 20: 984, 21: 772, 22: 609, 23: 205})])", | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8925498810177558, | |
| "eval_loss": 0.36934685707092285, | |
| "eval_runtime": 3.72, | |
| "eval_samples_per_second": 1468.551, | |
| "eval_steps_per_second": 11.559, | |
| "step": 6548 | |
| }, | |
| { | |
| "SuperNet": "OrderedDict([(<ElasticityDim.WIDTH: 'width'>, {0: 768, 1: 768, 2: 768, 3: 768, 4: 768, 5: 768, 6: 768, 7: 768, 8: 768, 9: 768, 10: 768, 11: 768, 12: 3072, 13: 3072, 14: 3072, 15: 3072, 16: 3072, 17: 3072, 18: 3072, 19: 3072, 20: 3072, 21: 3072, 22: 3072, 23: 3072})])", | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9200073219842577, | |
| "eval_loss": 0.28598278760910034, | |
| "eval_runtime": 4.4369, | |
| "eval_samples_per_second": 1231.274, | |
| "eval_steps_per_second": 9.692, | |
| "step": 6548 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 2.14, | |
| "learning_rate": 1.4372975696751451e-05, | |
| "loss": 0.3174, | |
| "step": 7000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 2.29, | |
| "learning_rate": 1.3640646425345818e-05, | |
| "loss": 0.298, | |
| "step": 7500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 2.44, | |
| "learning_rate": 1.2886582095197544e-05, | |
| "loss": 0.3068, | |
| "step": 8000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 2.6, | |
| "learning_rate": 1.2112581017647504e-05, | |
| "loss": 0.3003, | |
| "step": 8500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 2.75, | |
| "learning_rate": 1.1325079067048855e-05, | |
| "loss": 0.3005, | |
| "step": 9000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 2.9, | |
| "learning_rate": 1.0529108932691025e-05, | |
| "loss": 0.2972, | |
| "step": 9500 | |
| }, | |
| { | |
| "Minimum SubNet": "OrderedDict([(<ElasticityDim.WIDTH: 'width'>, {0: 512, 1: 320, 2: 320, 3: 576, 4: 576, 5: 512, 6: 448, 7: 448, 8: 448, 9: 320, 10: 384, 11: 448, 12: 1104, 13: 1066, 14: 1126, 15: 1102, 16: 1067, 17: 1023, 18: 1048, 19: 1061, 20: 984, 21: 772, 22: 609, 23: 205})])", | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8949295259015193, | |
| "eval_loss": 0.46985653042793274, | |
| "eval_runtime": 4.3758, | |
| "eval_samples_per_second": 1248.458, | |
| "eval_steps_per_second": 9.827, | |
| "step": 9822 | |
| }, | |
| { | |
| "SuperNet": "OrderedDict([(<ElasticityDim.WIDTH: 'width'>, {0: 768, 1: 768, 2: 768, 3: 768, 4: 768, 5: 768, 6: 768, 7: 768, 8: 768, 9: 768, 10: 768, 11: 768, 12: 3072, 13: 3072, 14: 3072, 15: 3072, 16: 3072, 17: 3072, 18: 3072, 19: 3072, 20: 3072, 21: 3072, 22: 3072, 23: 3072})])", | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9161632802489474, | |
| "eval_loss": 0.3910222053527832, | |
| "eval_runtime": 6.1209, | |
| "eval_samples_per_second": 892.518, | |
| "eval_steps_per_second": 7.025, | |
| "step": 9822 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 3.05, | |
| "learning_rate": 9.731356104066608e-06, | |
| "loss": 0.2217, | |
| "step": 10000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 3.21, | |
| "learning_rate": 8.933723081932598e-06, | |
| "loss": 0.1659, | |
| "step": 10500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 3.36, | |
| "learning_rate": 8.144475783733827e-06, | |
| "loss": 0.1606, | |
| "step": 11000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 3.51, | |
| "learning_rate": 7.365496426027108e-06, | |
| "loss": 0.1682, | |
| "step": 11500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 3.67, | |
| "learning_rate": 6.603353392229953e-06, | |
| "loss": 0.1692, | |
| "step": 12000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 3.82, | |
| "learning_rate": 5.8629173103837116e-06, | |
| "loss": 0.171, | |
| "step": 12500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 3.97, | |
| "learning_rate": 5.150318630440563e-06, | |
| "loss": 0.1611, | |
| "step": 13000 | |
| }, | |
| { | |
| "Minimum SubNet": "OrderedDict([(<ElasticityDim.WIDTH: 'width'>, {0: 512, 1: 320, 2: 320, 3: 576, 4: 576, 5: 512, 6: 448, 7: 448, 8: 448, 9: 320, 10: 384, 11: 448, 12: 1104, 13: 1066, 14: 1126, 15: 1102, 16: 1067, 17: 1023, 18: 1048, 19: 1061, 20: 984, 21: 772, 22: 609, 23: 205})])", | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8963939227530661, | |
| "eval_loss": 0.5541625618934631, | |
| "eval_runtime": 3.7938, | |
| "eval_samples_per_second": 1439.97, | |
| "eval_steps_per_second": 11.334, | |
| "step": 13096 | |
| }, | |
| { | |
| "SuperNet": "OrderedDict([(<ElasticityDim.WIDTH: 'width'>, {0: 768, 1: 768, 2: 768, 3: 768, 4: 768, 5: 768, 6: 768, 7: 768, 8: 768, 9: 768, 10: 768, 11: 768, 12: 3072, 13: 3072, 14: 3072, 15: 3072, 16: 3072, 17: 3072, 18: 3072, 19: 3072, 20: 3072, 21: 3072, 22: 3072, 23: 3072})])", | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9159802306425041, | |
| "eval_loss": 0.447282075881958, | |
| "eval_runtime": 4.4441, | |
| "eval_samples_per_second": 1229.257, | |
| "eval_steps_per_second": 9.676, | |
| "step": 13096 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 4.12, | |
| "learning_rate": 4.468589031016034e-06, | |
| "loss": 0.1269, | |
| "step": 13500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 4.28, | |
| "learning_rate": 3.820810355118921e-06, | |
| "loss": 0.1201, | |
| "step": 14000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 4.43, | |
| "learning_rate": 3.212521030987219e-06, | |
| "loss": 0.1193, | |
| "step": 14500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 4.58, | |
| "learning_rate": 2.6486925407131604e-06, | |
| "loss": 0.1117, | |
| "step": 15000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 4.73, | |
| "learning_rate": 2.1306695164823943e-06, | |
| "loss": 0.1133, | |
| "step": 15500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 4.89, | |
| "learning_rate": 1.6629370283058677e-06, | |
| "loss": 0.1155, | |
| "step": 16000 | |
| }, | |
| { | |
| "Minimum SubNet": "OrderedDict([(<ElasticityDim.WIDTH: 'width'>, {0: 512, 1: 320, 2: 320, 3: 576, 4: 576, 5: 512, 6: 448, 7: 448, 8: 448, 9: 320, 10: 384, 11: 448, 12: 1104, 13: 1066, 14: 1126, 15: 1102, 16: 1067, 17: 1023, 18: 1048, 19: 1061, 20: 984, 21: 772, 22: 609, 23: 205})])", | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8969430715723962, | |
| "eval_loss": 0.5925931334495544, | |
| "eval_runtime": 3.6303, | |
| "eval_samples_per_second": 1504.838, | |
| "eval_steps_per_second": 11.845, | |
| "step": 16370 | |
| }, | |
| { | |
| "SuperNet": "OrderedDict([(<ElasticityDim.WIDTH: 'width'>, {0: 768, 1: 768, 2: 768, 3: 768, 4: 768, 5: 768, 6: 768, 7: 768, 8: 768, 9: 768, 10: 768, 11: 768, 12: 3072, 13: 3072, 14: 3072, 15: 3072, 16: 3072, 17: 3072, 18: 3072, 19: 3072, 20: 3072, 21: 3072, 22: 3072, 23: 3072})])", | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.917993776313381, | |
| "eval_loss": 0.47876131534576416, | |
| "eval_runtime": 4.4184, | |
| "eval_samples_per_second": 1236.434, | |
| "eval_steps_per_second": 9.732, | |
| "step": 16370 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 5.04, | |
| "learning_rate": 1.2500322655233655e-06, | |
| "loss": 0.1131, | |
| "step": 16500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 5.19, | |
| "learning_rate": 8.91279249212037e-07, | |
| "loss": 0.0801, | |
| "step": 17000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 5.35, | |
| "learning_rate": 5.907373412389583e-07, | |
| "loss": 0.0911, | |
| "step": 17500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 5.5, | |
| "learning_rate": 3.5032721499437263e-07, | |
| "loss": 0.1044, | |
| "step": 18000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 5.65, | |
| "learning_rate": 1.7158525964722118e-07, | |
| "loss": 0.0938, | |
| "step": 18500 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 5.8, | |
| "learning_rate": 5.565376154176627e-08, | |
| "loss": 0.0962, | |
| "step": 19000 | |
| }, | |
| { | |
| "compression_loss": 0.0, | |
| "epoch": 5.96, | |
| "learning_rate": 3.2736041874104062e-09, | |
| "loss": 0.0867, | |
| "step": 19500 | |
| }, | |
| { | |
| "Minimum SubNet": "OrderedDict([(<ElasticityDim.WIDTH: 'width'>, {0: 512, 1: 320, 2: 320, 3: 576, 4: 576, 5: 512, 6: 448, 7: 448, 8: 448, 9: 320, 10: 384, 11: 448, 12: 1104, 13: 1066, 14: 1126, 15: 1102, 16: 1067, 17: 1023, 18: 1048, 19: 1061, 20: 984, 21: 772, 22: 609, 23: 205})])", | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8958447739337361, | |
| "eval_loss": 0.6002162098884583, | |
| "eval_runtime": 4.488, | |
| "eval_samples_per_second": 1217.259, | |
| "eval_steps_per_second": 9.581, | |
| "step": 19644 | |
| }, | |
| { | |
| "SuperNet": "OrderedDict([(<ElasticityDim.WIDTH: 'width'>, {0: 768, 1: 768, 2: 768, 3: 768, 4: 768, 5: 768, 6: 768, 7: 768, 8: 768, 9: 768, 10: 768, 11: 768, 12: 3072, 13: 3072, 14: 3072, 15: 3072, 16: 3072, 17: 3072, 18: 3072, 19: 3072, 20: 3072, 21: 3072, 22: 3072, 23: 3072})])", | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9176276771004942, | |
| "eval_loss": 0.4830775558948517, | |
| "eval_runtime": 6.2044, | |
| "eval_samples_per_second": 880.501, | |
| "eval_steps_per_second": 6.931, | |
| "step": 19644 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 19644, | |
| "total_flos": 4.133856190735872e+16, | |
| "train_loss": 0.3564307890654921, | |
| "train_runtime": 30220.0432, | |
| "train_samples_per_second": 20.796, | |
| "train_steps_per_second": 0.65 | |
| } | |
| ], | |
| "max_steps": 19644, | |
| "min_subnet_acc": null, | |
| "min_subnet_best_acc": null, | |
| "num_train_epochs": 6, | |
| "supernet_acc": null, | |
| "supernet_best_acc": null, | |
| "total_flos": 4.133856190735872e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |