| { | |
| "best_metric": 0.7853881278538812, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-piid/checkpoint-123", | |
| "epoch": 19.51219512195122, | |
| "eval_steps": 500, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.3499, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.2088, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_accuracy": 0.4520547945205479, | |
| "eval_loss": 1.1661090850830078, | |
| "eval_runtime": 1.4325, | |
| "eval_samples_per_second": 152.877, | |
| "eval_steps_per_second": 19.546, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.9539, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7545, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6073059360730594, | |
| "eval_loss": 0.8865624070167542, | |
| "eval_runtime": 1.3434, | |
| "eval_samples_per_second": 163.02, | |
| "eval_steps_per_second": 20.843, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.6405, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.6281, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_accuracy": 0.684931506849315, | |
| "eval_loss": 0.7788348197937012, | |
| "eval_runtime": 1.5459, | |
| "eval_samples_per_second": 141.667, | |
| "eval_steps_per_second": 18.113, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 0.5403, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.5939, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7397260273972602, | |
| "eval_loss": 0.644279420375824, | |
| "eval_runtime": 1.4721, | |
| "eval_samples_per_second": 148.762, | |
| "eval_steps_per_second": 19.02, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 4.305555555555556e-05, | |
| "loss": 0.5419, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.5254, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "eval_accuracy": 0.7808219178082192, | |
| "eval_loss": 0.5097259879112244, | |
| "eval_runtime": 1.552, | |
| "eval_samples_per_second": 141.107, | |
| "eval_steps_per_second": 18.041, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "learning_rate": 4.027777777777778e-05, | |
| "loss": 0.5295, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.5583, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7853881278538812, | |
| "eval_loss": 0.5714804530143738, | |
| "eval_runtime": 1.328, | |
| "eval_samples_per_second": 164.91, | |
| "eval_steps_per_second": 21.084, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4476, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.3463, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "eval_accuracy": 0.7351598173515982, | |
| "eval_loss": 0.6163173317909241, | |
| "eval_runtime": 1.4078, | |
| "eval_samples_per_second": 155.566, | |
| "eval_steps_per_second": 19.89, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 7.32, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.4396, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.3878, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7671232876712328, | |
| "eval_loss": 0.5671306848526001, | |
| "eval_runtime": 1.3088, | |
| "eval_samples_per_second": 167.328, | |
| "eval_steps_per_second": 21.394, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 3.194444444444444e-05, | |
| "loss": 0.3628, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.3653, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "eval_accuracy": 0.7579908675799086, | |
| "eval_loss": 0.5689936876296997, | |
| "eval_runtime": 1.3448, | |
| "eval_samples_per_second": 162.852, | |
| "eval_steps_per_second": 20.821, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 9.27, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 0.3083, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.3529, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7579908675799086, | |
| "eval_loss": 0.5939760208129883, | |
| "eval_runtime": 1.3291, | |
| "eval_samples_per_second": 164.77, | |
| "eval_steps_per_second": 21.067, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 10.24, | |
| "learning_rate": 2.6388888888888892e-05, | |
| "loss": 0.3382, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 10.73, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.301, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "eval_accuracy": 0.7625570776255708, | |
| "eval_loss": 0.6303488612174988, | |
| "eval_runtime": 1.3837, | |
| "eval_samples_per_second": 158.272, | |
| "eval_steps_per_second": 20.236, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 11.22, | |
| "learning_rate": 2.361111111111111e-05, | |
| "loss": 0.2672, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 11.71, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.2639, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.776255707762557, | |
| "eval_loss": 0.5725071430206299, | |
| "eval_runtime": 1.3678, | |
| "eval_samples_per_second": 160.117, | |
| "eval_steps_per_second": 20.472, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 12.2, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.2627, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 12.68, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.2847, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 12.98, | |
| "eval_accuracy": 0.771689497716895, | |
| "eval_loss": 0.6279605031013489, | |
| "eval_runtime": 1.3676, | |
| "eval_samples_per_second": 160.133, | |
| "eval_steps_per_second": 20.474, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "learning_rate": 1.8055555555555555e-05, | |
| "loss": 0.2534, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.25, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.771689497716895, | |
| "eval_loss": 0.5975117683410645, | |
| "eval_runtime": 1.3451, | |
| "eval_samples_per_second": 162.816, | |
| "eval_steps_per_second": 20.817, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "learning_rate": 1.527777777777778e-05, | |
| "loss": 0.3332, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.2472, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "eval_accuracy": 0.7671232876712328, | |
| "eval_loss": 0.5821006894111633, | |
| "eval_runtime": 1.3394, | |
| "eval_samples_per_second": 163.506, | |
| "eval_steps_per_second": 20.905, | |
| "step": 307 | |
| }, | |
| { | |
| "epoch": 15.12, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2185, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 15.61, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.1676, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7625570776255708, | |
| "eval_loss": 0.6456273198127747, | |
| "eval_runtime": 1.3379, | |
| "eval_samples_per_second": 163.684, | |
| "eval_steps_per_second": 20.928, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 16.1, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.1898, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 16.59, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.1327, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "eval_accuracy": 0.7671232876712328, | |
| "eval_loss": 0.6117277145385742, | |
| "eval_runtime": 1.3936, | |
| "eval_samples_per_second": 157.15, | |
| "eval_steps_per_second": 20.092, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 17.07, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.2238, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 17.56, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.1977, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7488584474885844, | |
| "eval_loss": 0.6987997889518738, | |
| "eval_runtime": 1.3339, | |
| "eval_samples_per_second": 164.179, | |
| "eval_steps_per_second": 20.991, | |
| "step": 369 | |
| }, | |
| { | |
| "epoch": 18.05, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.3006, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 18.54, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.1602, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "eval_accuracy": 0.7671232876712328, | |
| "eval_loss": 0.6448192000389099, | |
| "eval_runtime": 1.3619, | |
| "eval_samples_per_second": 160.799, | |
| "eval_steps_per_second": 20.559, | |
| "step": 389 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "learning_rate": 1.388888888888889e-06, | |
| "loss": 0.231, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 19.51, | |
| "learning_rate": 0.0, | |
| "loss": 0.1785, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.51, | |
| "eval_accuracy": 0.771689497716895, | |
| "eval_loss": 0.633254885673523, | |
| "eval_runtime": 1.3431, | |
| "eval_samples_per_second": 163.06, | |
| "eval_steps_per_second": 20.848, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.51, | |
| "step": 400, | |
| "total_flos": 3.167575685961523e+17, | |
| "train_loss": 0.4159336131811142, | |
| "train_runtime": 271.0168, | |
| "train_samples_per_second": 48.189, | |
| "train_steps_per_second": 1.476 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 400, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 3.167575685961523e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |