| { | |
| "best_metric": 0.08002225309610367, | |
| "best_model_checkpoint": "./vit-base-nthu-ddd/checkpoint-6000", | |
| "epoch": 2.0, | |
| "global_step": 6652, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.00019702345159350573, | |
| "loss": 0.1091, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00019404690318701145, | |
| "loss": 0.1244, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00019104028863499702, | |
| "loss": 0.114, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00018803367408298257, | |
| "loss": 0.1259, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00018502705953096814, | |
| "loss": 0.1257, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0001820204449789537, | |
| "loss": 0.1064, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0001790438965724594, | |
| "loss": 0.1172, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00017603728202044498, | |
| "loss": 0.1186, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00017303066746843056, | |
| "loss": 0.0997, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00017002405291641613, | |
| "loss": 0.1223, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00016701743836440168, | |
| "loss": 0.1081, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00016401082381238725, | |
| "loss": 0.1167, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00016100420926037283, | |
| "loss": 0.0867, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0001579975947083584, | |
| "loss": 0.1048, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00015499098015634398, | |
| "loss": 0.1142, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00015198436560432953, | |
| "loss": 0.1033, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00014900781719783524, | |
| "loss": 0.1123, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00014600120264582082, | |
| "loss": 0.1353, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00014299458809380637, | |
| "loss": 0.0924, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00013998797354179194, | |
| "loss": 0.1173, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00013698135898977752, | |
| "loss": 0.0863, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0001339747444377631, | |
| "loss": 0.1035, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00013096812988574864, | |
| "loss": 0.0897, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0001279615153337342, | |
| "loss": 0.0943, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.0001249549007817198, | |
| "loss": 0.1159, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00012194828622970536, | |
| "loss": 0.0931, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00011894167167769092, | |
| "loss": 0.1131, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0001159350571256765, | |
| "loss": 0.1052, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00011292844257366206, | |
| "loss": 0.0944, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.00010992182802164763, | |
| "loss": 0.0873, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.9549793310785419, | |
| "eval_loss": 0.1255834847688675, | |
| "eval_runtime": 240.5647, | |
| "eval_samples_per_second": 55.307, | |
| "eval_steps_per_second": 6.917, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.00010691521346963321, | |
| "loss": 0.0907, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00010390859891761877, | |
| "loss": 0.0836, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.00010090198436560434, | |
| "loss": 0.0791, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 9.78953698135899e-05, | |
| "loss": 0.0953, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.488875526157547e-05, | |
| "loss": 0.0859, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 9.191220685508118e-05, | |
| "loss": 0.1059, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 8.890559230306676e-05, | |
| "loss": 0.0859, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 8.589897775105232e-05, | |
| "loss": 0.0965, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 8.28923631990379e-05, | |
| "loss": 0.0801, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 7.988574864702346e-05, | |
| "loss": 0.1072, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 7.687913409500902e-05, | |
| "loss": 0.0771, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 7.387251954299459e-05, | |
| "loss": 0.0854, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 7.086590499098017e-05, | |
| "loss": 0.0721, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 6.785929043896573e-05, | |
| "loss": 0.0929, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 6.48526758869513e-05, | |
| "loss": 0.0649, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 6.184606133493686e-05, | |
| "loss": 0.085, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 5.883944678292244e-05, | |
| "loss": 0.0838, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 5.5832832230908004e-05, | |
| "loss": 0.0629, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 5.2826217678893566e-05, | |
| "loss": 0.0985, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 4.981960312687914e-05, | |
| "loss": 0.0656, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 4.68129885748647e-05, | |
| "loss": 0.0814, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 4.380637402285027e-05, | |
| "loss": 0.0667, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 4.0799759470835844e-05, | |
| "loss": 0.0944, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 3.779314491882141e-05, | |
| "loss": 0.0566, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 3.478653036680698e-05, | |
| "loss": 0.0867, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 3.177991581479254e-05, | |
| "loss": 0.0705, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 2.877330126277811e-05, | |
| "loss": 0.0734, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.5766686710763683e-05, | |
| "loss": 0.0781, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 2.2760072158749247e-05, | |
| "loss": 0.0712, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.975345760673482e-05, | |
| "loss": 0.06, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_accuracy": 0.9751972942502819, | |
| "eval_loss": 0.08002225309610367, | |
| "eval_runtime": 240.0279, | |
| "eval_samples_per_second": 55.431, | |
| "eval_steps_per_second": 6.933, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 1.6746843054720386e-05, | |
| "loss": 0.0698, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 1.3740228502705954e-05, | |
| "loss": 0.0705, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 1.0733613950691522e-05, | |
| "loss": 0.0553, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 7.72699939867709e-06, | |
| "loss": 0.0743, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 4.750450992182803e-06, | |
| "loss": 0.0636, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 1.7438364401683706e-06, | |
| "loss": 0.083, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 6652, | |
| "total_flos": 8.247627438660256e+18, | |
| "train_loss": 0.09209114413482199, | |
| "train_runtime": 3805.7384, | |
| "train_samples_per_second": 27.966, | |
| "train_steps_per_second": 1.748 | |
| } | |
| ], | |
| "max_steps": 6652, | |
| "num_train_epochs": 2, | |
| "total_flos": 8.247627438660256e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |