| { | |
| "best_metric": 0.872196261682243, | |
| "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-cassava/checkpoint-399", | |
| "epoch": 2.994392523364486, | |
| "global_step": 399, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.6059, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.381, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.1472, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0363, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 4.860724233983287e-05, | |
| "loss": 0.9067, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.721448467966574e-05, | |
| "loss": 0.7733, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.582172701949861e-05, | |
| "loss": 0.7452, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.442896935933148e-05, | |
| "loss": 0.6585, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.303621169916435e-05, | |
| "loss": 0.6018, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.164345403899722e-05, | |
| "loss": 0.5398, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.0250696378830085e-05, | |
| "loss": 0.5506, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 3.885793871866296e-05, | |
| "loss": 0.5615, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 3.746518105849583e-05, | |
| "loss": 0.5242, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.8502336448598131, | |
| "eval_loss": 0.4905190169811249, | |
| "eval_runtime": 89.5244, | |
| "eval_samples_per_second": 47.808, | |
| "eval_steps_per_second": 1.497, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 3.607242339832869e-05, | |
| "loss": 0.5078, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 3.4679665738161556e-05, | |
| "loss": 0.5059, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3.328690807799443e-05, | |
| "loss": 0.4914, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 3.18941504178273e-05, | |
| "loss": 0.4694, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 3.050139275766017e-05, | |
| "loss": 0.4538, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 2.9108635097493035e-05, | |
| "loss": 0.4467, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 2.7715877437325903e-05, | |
| "loss": 0.4958, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 2.6323119777158778e-05, | |
| "loss": 0.4647, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 2.4930362116991646e-05, | |
| "loss": 0.413, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 2.3537604456824514e-05, | |
| "loss": 0.4005, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 2.214484679665738e-05, | |
| "loss": 0.4374, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 2.0752089136490253e-05, | |
| "loss": 0.4366, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 1.935933147632312e-05, | |
| "loss": 0.4405, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.8696261682242991, | |
| "eval_loss": 0.41539567708969116, | |
| "eval_runtime": 92.6526, | |
| "eval_samples_per_second": 46.194, | |
| "eval_steps_per_second": 1.446, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 1.7966573816155992e-05, | |
| "loss": 0.4296, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 1.6573816155988857e-05, | |
| "loss": 0.4433, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 1.518105849582173e-05, | |
| "loss": 0.4323, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 1.3788300835654596e-05, | |
| "loss": 0.4169, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 1.2395543175487466e-05, | |
| "loss": 0.395, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1.1002785515320335e-05, | |
| "loss": 0.4045, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 9.610027855153205e-06, | |
| "loss": 0.3718, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 8.217270194986073e-06, | |
| "loss": 0.4601, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 6.8245125348189415e-06, | |
| "loss": 0.3969, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 5.43175487465181e-06, | |
| "loss": 0.397, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 4.03899721448468e-06, | |
| "loss": 0.3909, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 2.6462395543175487e-06, | |
| "loss": 0.4032, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 1.253481894150418e-06, | |
| "loss": 0.3938, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.872196261682243, | |
| "eval_loss": 0.4010835587978363, | |
| "eval_runtime": 89.5202, | |
| "eval_samples_per_second": 47.81, | |
| "eval_steps_per_second": 1.497, | |
| "step": 399 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "step": 399, | |
| "total_flos": 3.9721912052731453e+18, | |
| "train_loss": 0.5681018877148927, | |
| "train_runtime": 3778.1283, | |
| "train_samples_per_second": 13.592, | |
| "train_steps_per_second": 0.106 | |
| } | |
| ], | |
| "max_steps": 399, | |
| "num_train_epochs": 3, | |
| "total_flos": 3.9721912052731453e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |