{ "best_metric": null, "best_model_checkpoint": null, "best_supernet_model_checkpoint": null, "epoch": 4.0, "global_step": 8420, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "compression_loss": 0.0, "epoch": 0.24, "learning_rate": 2.9743870065874887e-05, "loss": 0.5172, "step": 500 }, { "compression_loss": 0.0, "epoch": 0.48, "learning_rate": 2.8976113714195393e-05, "loss": 0.2933, "step": 1000 }, { "compression_loss": 0.0, "epoch": 0.71, "learning_rate": 2.7723356990465294e-05, "loss": 0.216, "step": 1500 }, { "compression_loss": 0.0, "epoch": 0.95, "learning_rate": 2.6029073172701843e-05, "loss": 0.1786, "step": 2000 }, { "Minimum SubNet": "OrderedDict([(, {0: 320, 1: 256, 2: 256, 3: 192, 4: 256, 5: 192, 6: 128, 7: 320, 8: 256, 9: 192, 10: 128, 11: 64, 12: 1585, 13: 1570, 14: 1775, 15: 1717, 16: 1679, 17: 1580, 18: 1621, 19: 1406, 20: 1188, 21: 930, 22: 828, 23: 654})])", "epoch": 1.0, "eval_accuracy": 0.9151376146788991, "eval_loss": 0.31559914350509644, "eval_runtime": 0.6902, "eval_samples_per_second": 1263.387, "eval_steps_per_second": 10.142, "step": 2105 }, { "SuperNet": "OrderedDict([(, {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": 0.9220183486238532, "eval_loss": 0.2937995493412018, "eval_runtime": 1.0139, "eval_samples_per_second": 860.065, "eval_steps_per_second": 6.904, "step": 2105 }, { "compression_loss": 0.0, "epoch": 1.19, "learning_rate": 2.395654751860746e-05, "loss": 0.1126, "step": 2500 }, { "compression_loss": 0.0, "epoch": 1.43, "learning_rate": 2.1569418439843677e-05, "loss": 0.0894, "step": 3000 }, { "compression_loss": 0.0, "epoch": 1.66, "learning_rate": 1.895431680356957e-05, "loss": 0.0786, "step": 3500 }, { "compression_loss": 0.0, "epoch": 1.9, "learning_rate": 1.6201992105136965e-05, "loss": 0.0868, "step": 4000 }, { "Minimum SubNet": "OrderedDict([(, {0: 320, 1: 256, 2: 256, 3: 192, 4: 256, 5: 192, 6: 128, 7: 320, 8: 256, 9: 192, 10: 128, 11: 64, 12: 1585, 13: 1570, 14: 1775, 15: 1717, 16: 1679, 17: 1580, 18: 1621, 19: 1406, 20: 1188, 21: 930, 22: 828, 23: 654})])", "epoch": 2.0, "eval_accuracy": 0.9197247706422018, "eval_loss": 0.3034738302230835, "eval_runtime": 0.6917, "eval_samples_per_second": 1260.72, "eval_steps_per_second": 10.12, "step": 4210 }, { "SuperNet": "OrderedDict([(, {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": 0.9231651376146789, "eval_loss": 0.3007892668247223, "eval_runtime": 1.0119, "eval_samples_per_second": 861.755, "eval_steps_per_second": 6.918, "step": 4210 }, { "compression_loss": 0.0, "epoch": 2.14, "learning_rate": 1.3413520927067597e-05, "loss": 0.0576, "step": 4500 }, { "compression_loss": 0.0, "epoch": 2.38, "learning_rate": 1.0674525310461774e-05, "loss": 0.0435, "step": 5000 }, { "compression_loss": 0.0, "epoch": 2.61, "learning_rate": 8.085632710804847e-06, "loss": 0.0411, "step": 5500 }, { "compression_loss": 0.0, "epoch": 2.85, "learning_rate": 5.736683115070964e-06, "loss": 0.0371, "step": 6000 }, { "Minimum SubNet": "OrderedDict([(, {0: 320, 1: 256, 2: 256, 3: 192, 4: 256, 5: 192, 6: 128, 7: 320, 8: 256, 9: 192, 10: 128, 11: 64, 12: 1585, 13: 1570, 14: 1775, 15: 1717, 16: 1679, 17: 1580, 18: 1621, 19: 1406, 20: 1188, 21: 930, 22: 828, 23: 654})])", "epoch": 3.0, "eval_accuracy": 0.9151376146788991, "eval_loss": 0.30734896659851074, "eval_runtime": 0.7135, "eval_samples_per_second": 1222.206, "eval_steps_per_second": 9.811, "step": 6315 }, { "SuperNet": "OrderedDict([(, {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": 0.9288990825688074, "eval_loss": 0.26742127537727356, "eval_runtime": 1.0141, "eval_samples_per_second": 859.88, "eval_steps_per_second": 6.903, "step": 6315 }, { "compression_loss": 0.0, "epoch": 3.09, "learning_rate": 3.7128752783657687e-06, "loss": 0.0341, "step": 6500 }, { "compression_loss": 0.0, "epoch": 3.33, "learning_rate": 2.0791936159197827e-06, "loss": 0.0285, "step": 7000 }, { "compression_loss": 0.0, "epoch": 3.56, "learning_rate": 8.902040909382414e-07, "loss": 0.0253, "step": 7500 }, { "compression_loss": 0.0, "epoch": 3.8, "learning_rate": 1.9085398971886657e-07, "loss": 0.0249, "step": 8000 }, { "Minimum SubNet": "OrderedDict([(, {0: 320, 1: 256, 2: 256, 3: 192, 4: 256, 5: 192, 6: 128, 7: 320, 8: 256, 9: 192, 10: 128, 11: 64, 12: 1585, 13: 1570, 14: 1775, 15: 1717, 16: 1679, 17: 1580, 18: 1621, 19: 1406, 20: 1188, 21: 930, 22: 828, 23: 654})])", "epoch": 4.0, "eval_accuracy": 0.9139908256880734, "eval_loss": 0.3040490746498108, "eval_runtime": 0.7023, "eval_samples_per_second": 1241.664, "eval_steps_per_second": 9.967, "step": 8420 }, { "SuperNet": "OrderedDict([(, {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": 0.9311926605504587, "eval_loss": 0.26580190658569336, "eval_runtime": 1.0433, "eval_samples_per_second": 835.844, "eval_steps_per_second": 6.71, "step": 8420 }, { "epoch": 4.0, "step": 8420, "total_flos": 1.772026646744064e+16, "train_loss": 0.11200937262057126, "train_runtime": 14148.7677, "train_samples_per_second": 19.04, "train_steps_per_second": 0.595 } ], "max_steps": 8420, "min_subnet_acc": null, "min_subnet_best_acc": null, "num_train_epochs": 4, "supernet_acc": null, "supernet_best_acc": null, "total_flos": 1.772026646744064e+16, "trial_name": null, "trial_params": null }