| { | |
| "best_metric": 0.9559725730783111, | |
| "best_model_checkpoint": "violation-classification-bantai-vit-v80ep/checkpoint-8000", | |
| "epoch": 79.99753086419753, | |
| "global_step": 8080, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 3.094059405940594e-05, | |
| "loss": 0.797, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "eval_accuracy": 0.8715265247203176, | |
| "eval_loss": 0.392585426568985, | |
| "eval_runtime": 63.7974, | |
| "eval_samples_per_second": 86.869, | |
| "eval_steps_per_second": 2.727, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 4.867986798679868e-05, | |
| "loss": 0.3095, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "eval_accuracy": 0.9106820642367376, | |
| "eval_loss": 0.2596624493598938, | |
| "eval_runtime": 63.8682, | |
| "eval_samples_per_second": 86.772, | |
| "eval_steps_per_second": 2.724, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 14.85, | |
| "learning_rate": 4.5242024202420245e-05, | |
| "loss": 0.1726, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 14.85, | |
| "eval_accuracy": 0.9252977264525442, | |
| "eval_loss": 0.2156512439250946, | |
| "eval_runtime": 63.7151, | |
| "eval_samples_per_second": 86.981, | |
| "eval_steps_per_second": 2.731, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 19.8, | |
| "learning_rate": 4.18041804180418e-05, | |
| "loss": 0.1259, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 19.8, | |
| "eval_accuracy": 0.939191627571274, | |
| "eval_loss": 0.1869734823703766, | |
| "eval_runtime": 63.965, | |
| "eval_samples_per_second": 86.641, | |
| "eval_steps_per_second": 2.72, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 24.75, | |
| "learning_rate": 3.8366336633663367e-05, | |
| "loss": 0.0959, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 24.75, | |
| "eval_accuracy": 0.9444243955250812, | |
| "eval_loss": 0.17974421381950378, | |
| "eval_runtime": 63.8189, | |
| "eval_samples_per_second": 86.84, | |
| "eval_steps_per_second": 2.726, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 29.7, | |
| "learning_rate": 3.492849284928493e-05, | |
| "loss": 0.0835, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 29.7, | |
| "eval_accuracy": 0.9354023818116204, | |
| "eval_loss": 0.22931958734989166, | |
| "eval_runtime": 63.8202, | |
| "eval_samples_per_second": 86.838, | |
| "eval_steps_per_second": 2.726, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 34.65, | |
| "learning_rate": 3.149064906490649e-05, | |
| "loss": 0.0722, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 34.65, | |
| "eval_accuracy": 0.9440635149765427, | |
| "eval_loss": 0.19207924604415894, | |
| "eval_runtime": 63.813, | |
| "eval_samples_per_second": 86.848, | |
| "eval_steps_per_second": 2.727, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 39.6, | |
| "learning_rate": 2.8052805280528056e-05, | |
| "loss": 0.0628, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 39.6, | |
| "eval_accuracy": 0.9491158426560808, | |
| "eval_loss": 0.18968293070793152, | |
| "eval_runtime": 63.6858, | |
| "eval_samples_per_second": 87.021, | |
| "eval_steps_per_second": 2.732, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 44.55, | |
| "learning_rate": 2.4614961496149617e-05, | |
| "loss": 0.059, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 44.55, | |
| "eval_accuracy": 0.9520028870443883, | |
| "eval_loss": 0.1719067543745041, | |
| "eval_runtime": 63.8275, | |
| "eval_samples_per_second": 86.828, | |
| "eval_steps_per_second": 2.726, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 49.5, | |
| "learning_rate": 2.1177117711771178e-05, | |
| "loss": 0.0531, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 49.5, | |
| "eval_accuracy": 0.9512811259473114, | |
| "eval_loss": 0.19867293536663055, | |
| "eval_runtime": 63.7873, | |
| "eval_samples_per_second": 86.882, | |
| "eval_steps_per_second": 2.728, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 54.45, | |
| "learning_rate": 1.7739273927392743e-05, | |
| "loss": 0.046, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 54.45, | |
| "eval_accuracy": 0.9556116925297726, | |
| "eval_loss": 0.17134784162044525, | |
| "eval_runtime": 63.8123, | |
| "eval_samples_per_second": 86.848, | |
| "eval_steps_per_second": 2.727, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 59.4, | |
| "learning_rate": 1.4301430143014302e-05, | |
| "loss": 0.0444, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 59.4, | |
| "eval_accuracy": 0.952544207867196, | |
| "eval_loss": 0.2016247808933258, | |
| "eval_runtime": 63.9341, | |
| "eval_samples_per_second": 86.683, | |
| "eval_steps_per_second": 2.722, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 64.36, | |
| "learning_rate": 1.0863586358635865e-05, | |
| "loss": 0.042, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 64.36, | |
| "eval_accuracy": 0.952544207867196, | |
| "eval_loss": 0.19500622153282166, | |
| "eval_runtime": 63.6967, | |
| "eval_samples_per_second": 87.006, | |
| "eval_steps_per_second": 2.732, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 69.31, | |
| "learning_rate": 7.4257425742574256e-06, | |
| "loss": 0.0363, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 69.31, | |
| "eval_accuracy": 0.9548899314326957, | |
| "eval_loss": 0.20167508721351624, | |
| "eval_runtime": 63.6587, | |
| "eval_samples_per_second": 87.058, | |
| "eval_steps_per_second": 2.733, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 74.26, | |
| "learning_rate": 3.987898789878988e-06, | |
| "loss": 0.037, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 74.26, | |
| "eval_accuracy": 0.955070371706965, | |
| "eval_loss": 0.1943303495645523, | |
| "eval_runtime": 63.8065, | |
| "eval_samples_per_second": 86.856, | |
| "eval_steps_per_second": 2.727, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 79.21, | |
| "learning_rate": 5.500550055005501e-07, | |
| "loss": 0.0343, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 79.21, | |
| "eval_accuracy": 0.9559725730783111, | |
| "eval_loss": 0.19736222922801971, | |
| "eval_runtime": 63.8534, | |
| "eval_samples_per_second": 86.793, | |
| "eval_steps_per_second": 2.725, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "step": 8080, | |
| "total_flos": 8.01589959270506e+19, | |
| "train_loss": 0.12853888580114534, | |
| "train_runtime": 24615.3911, | |
| "train_samples_per_second": 42.022, | |
| "train_steps_per_second": 0.328 | |
| } | |
| ], | |
| "max_steps": 8080, | |
| "num_train_epochs": 80, | |
| "total_flos": 8.01589959270506e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |