| { | |
| "best_metric": 0.42657342657342656, | |
| "best_model_checkpoint": "vit-base-patch16-224-for-pre_evaluation/checkpoint-236", | |
| "epoch": 27.906976744186046, | |
| "eval_steps": 500, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.6266, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_accuracy": 0.3006993006993007, | |
| "eval_loss": 1.5307577848434448, | |
| "eval_runtime": 10.578, | |
| "eval_samples_per_second": 40.556, | |
| "eval_steps_per_second": 1.324, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.4973, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_accuracy": 0.3123543123543124, | |
| "eval_loss": 1.5000418424606323, | |
| "eval_runtime": 7.5932, | |
| "eval_samples_per_second": 56.498, | |
| "eval_steps_per_second": 1.844, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4636, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_accuracy": 0.34498834498834496, | |
| "eval_loss": 1.472951054573059, | |
| "eval_runtime": 11.6515, | |
| "eval_samples_per_second": 36.819, | |
| "eval_steps_per_second": 1.202, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 1.4185, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.30536130536130535, | |
| "eval_loss": 1.5051661729812622, | |
| "eval_runtime": 7.0051, | |
| "eval_samples_per_second": 61.241, | |
| "eval_steps_per_second": 1.999, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 1.4147, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "eval_accuracy": 0.34965034965034963, | |
| "eval_loss": 1.4455597400665283, | |
| "eval_runtime": 6.741, | |
| "eval_samples_per_second": 63.641, | |
| "eval_steps_per_second": 2.077, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 1.337, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.95, | |
| "eval_accuracy": 0.34265734265734266, | |
| "eval_loss": 1.4735510349273682, | |
| "eval_runtime": 7.0594, | |
| "eval_samples_per_second": 60.77, | |
| "eval_steps_per_second": 1.983, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 1.2869, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "eval_accuracy": 0.37995337995337997, | |
| "eval_loss": 1.4170001745224, | |
| "eval_runtime": 7.3984, | |
| "eval_samples_per_second": 57.985, | |
| "eval_steps_per_second": 1.892, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 7.44, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 1.241, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.372960372960373, | |
| "eval_loss": 1.5077800750732422, | |
| "eval_runtime": 7.0787, | |
| "eval_samples_per_second": 60.604, | |
| "eval_steps_per_second": 1.978, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 8.37, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 1.2261, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "eval_accuracy": 0.38927738927738925, | |
| "eval_loss": 1.4629043340682983, | |
| "eval_runtime": 7.4136, | |
| "eval_samples_per_second": 57.867, | |
| "eval_steps_per_second": 1.888, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 1.1967, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "eval_accuracy": 0.41025641025641024, | |
| "eval_loss": 1.4139227867126465, | |
| "eval_runtime": 7.4631, | |
| "eval_samples_per_second": 57.483, | |
| "eval_steps_per_second": 1.876, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 10.23, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 1.1153, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "eval_accuracy": 0.4195804195804196, | |
| "eval_loss": 1.4337666034698486, | |
| "eval_runtime": 7.0965, | |
| "eval_samples_per_second": 60.452, | |
| "eval_steps_per_second": 1.973, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 11.16, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.063, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.3939393939393939, | |
| "eval_loss": 1.4642902612686157, | |
| "eval_runtime": 7.3068, | |
| "eval_samples_per_second": 58.713, | |
| "eval_steps_per_second": 1.916, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 12.09, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 1.0434, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 12.93, | |
| "eval_accuracy": 0.40792540792540793, | |
| "eval_loss": 1.4725754261016846, | |
| "eval_runtime": 7.4877, | |
| "eval_samples_per_second": 57.294, | |
| "eval_steps_per_second": 1.87, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 1.0067, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 13.95, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.9849, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 13.95, | |
| "eval_accuracy": 0.4149184149184149, | |
| "eval_loss": 1.528671145439148, | |
| "eval_runtime": 7.2601, | |
| "eval_samples_per_second": 59.09, | |
| "eval_steps_per_second": 1.928, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 14.88, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.9285, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "eval_accuracy": 0.3986013986013986, | |
| "eval_loss": 1.552985429763794, | |
| "eval_runtime": 7.0224, | |
| "eval_samples_per_second": 61.09, | |
| "eval_steps_per_second": 1.994, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 15.81, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.8724, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.3939393939393939, | |
| "eval_loss": 1.5016406774520874, | |
| "eval_runtime": 7.1273, | |
| "eval_samples_per_second": 60.191, | |
| "eval_steps_per_second": 1.964, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 16.74, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.8063, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 16.93, | |
| "eval_accuracy": 0.40559440559440557, | |
| "eval_loss": 1.5184847116470337, | |
| "eval_runtime": 7.1223, | |
| "eval_samples_per_second": 60.233, | |
| "eval_steps_per_second": 1.966, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 17.67, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.8206, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 17.95, | |
| "eval_accuracy": 0.4172494172494173, | |
| "eval_loss": 1.5447591543197632, | |
| "eval_runtime": 7.0975, | |
| "eval_samples_per_second": 60.444, | |
| "eval_steps_per_second": 1.973, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 18.6, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.7396, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "eval_accuracy": 0.40326340326340326, | |
| "eval_loss": 1.554015874862671, | |
| "eval_runtime": 6.78, | |
| "eval_samples_per_second": 63.275, | |
| "eval_steps_per_second": 2.065, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 19.53, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.7437, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.40093240093240096, | |
| "eval_loss": 1.579628586769104, | |
| "eval_runtime": 6.8036, | |
| "eval_samples_per_second": 63.055, | |
| "eval_steps_per_second": 2.058, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 20.47, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.7103, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 20.93, | |
| "eval_accuracy": 0.40326340326340326, | |
| "eval_loss": 1.6218039989471436, | |
| "eval_runtime": 7.5951, | |
| "eval_samples_per_second": 56.484, | |
| "eval_steps_per_second": 1.843, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 21.4, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.6861, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 21.95, | |
| "eval_accuracy": 0.42657342657342656, | |
| "eval_loss": 1.6125619411468506, | |
| "eval_runtime": 7.0894, | |
| "eval_samples_per_second": 60.512, | |
| "eval_steps_per_second": 1.975, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 22.33, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.6798, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 22.98, | |
| "eval_accuracy": 0.42657342657342656, | |
| "eval_loss": 1.6051045656204224, | |
| "eval_runtime": 7.0114, | |
| "eval_samples_per_second": 61.186, | |
| "eval_steps_per_second": 1.997, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 23.26, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.6358, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.41025641025641024, | |
| "eval_loss": 1.6141223907470703, | |
| "eval_runtime": 6.9962, | |
| "eval_samples_per_second": 61.319, | |
| "eval_steps_per_second": 2.001, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 24.19, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.6189, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 24.93, | |
| "eval_accuracy": 0.40559440559440557, | |
| "eval_loss": 1.6705572605133057, | |
| "eval_runtime": 6.7024, | |
| "eval_samples_per_second": 64.007, | |
| "eval_steps_per_second": 2.089, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 25.12, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.6176, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 25.95, | |
| "eval_accuracy": 0.4219114219114219, | |
| "eval_loss": 1.6404389142990112, | |
| "eval_runtime": 6.7355, | |
| "eval_samples_per_second": 63.692, | |
| "eval_steps_per_second": 2.079, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 26.05, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.575, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 26.98, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.5756, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 26.98, | |
| "eval_accuracy": 0.4172494172494173, | |
| "eval_loss": 1.660282850265503, | |
| "eval_runtime": 6.8437, | |
| "eval_samples_per_second": 62.685, | |
| "eval_steps_per_second": 2.046, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 27.91, | |
| "learning_rate": 0.0, | |
| "loss": 0.5887, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 27.91, | |
| "eval_accuracy": 0.42424242424242425, | |
| "eval_loss": 1.6626534461975098, | |
| "eval_runtime": 6.7064, | |
| "eval_samples_per_second": 63.969, | |
| "eval_steps_per_second": 2.088, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 27.91, | |
| "step": 300, | |
| "total_flos": 2.9422949721372426e+18, | |
| "train_loss": 0.9840216811498006, | |
| "train_runtime": 1797.847, | |
| "train_samples_per_second": 22.694, | |
| "train_steps_per_second": 0.167 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 300, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 2.9422949721372426e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |