| { | |
| "best_metric": 0.8314176245210728, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-vit0/checkpoint-296", | |
| "epoch": 19.45945945945946, | |
| "eval_steps": 500, | |
| "global_step": 360, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 1.13, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_accuracy": 0.4329501915708812, | |
| "eval_loss": 1.0297268629074097, | |
| "eval_runtime": 147.8115, | |
| "eval_samples_per_second": 1.766, | |
| "eval_steps_per_second": 0.061, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 1.0527, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.9066, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6590038314176245, | |
| "eval_loss": 0.8349204659461975, | |
| "eval_runtime": 13.2219, | |
| "eval_samples_per_second": 19.74, | |
| "eval_steps_per_second": 0.681, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 4.938271604938271e-05, | |
| "loss": 0.8065, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 4.783950617283951e-05, | |
| "loss": 0.7157, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.6743295019157088, | |
| "eval_loss": 0.8049836754798889, | |
| "eval_runtime": 13.1267, | |
| "eval_samples_per_second": 19.883, | |
| "eval_steps_per_second": 0.686, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.6831, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 4.4753086419753084e-05, | |
| "loss": 0.6446, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7164750957854407, | |
| "eval_loss": 0.6934032440185547, | |
| "eval_runtime": 13.1723, | |
| "eval_samples_per_second": 19.814, | |
| "eval_steps_per_second": 0.683, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 4.3209876543209875e-05, | |
| "loss": 0.5756, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.5707, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "eval_accuracy": 0.7432950191570882, | |
| "eval_loss": 0.6323675513267517, | |
| "eval_runtime": 13.1405, | |
| "eval_samples_per_second": 19.862, | |
| "eval_steps_per_second": 0.685, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 4.012345679012346e-05, | |
| "loss": 0.5341, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.95, | |
| "learning_rate": 3.8580246913580246e-05, | |
| "loss": 0.5042, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.735632183908046, | |
| "eval_loss": 0.6155727505683899, | |
| "eval_runtime": 12.7114, | |
| "eval_samples_per_second": 20.533, | |
| "eval_steps_per_second": 0.708, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.4714, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.6825360655784607, | |
| "eval_runtime": 13.3632, | |
| "eval_samples_per_second": 19.531, | |
| "eval_steps_per_second": 0.673, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 3.5493827160493834e-05, | |
| "loss": 0.4471, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 7.57, | |
| "learning_rate": 3.395061728395062e-05, | |
| "loss": 0.4225, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7624521072796935, | |
| "eval_loss": 0.5692493915557861, | |
| "eval_runtime": 12.4031, | |
| "eval_samples_per_second": 21.043, | |
| "eval_steps_per_second": 0.726, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 8.11, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.4066, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.65, | |
| "learning_rate": 3.08641975308642e-05, | |
| "loss": 0.3912, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "eval_accuracy": 0.7586206896551724, | |
| "eval_loss": 0.615011990070343, | |
| "eval_runtime": 13.0398, | |
| "eval_samples_per_second": 20.016, | |
| "eval_steps_per_second": 0.69, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "learning_rate": 2.9320987654320992e-05, | |
| "loss": 0.3342, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.3442, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8007662835249042, | |
| "eval_loss": 0.4901227355003357, | |
| "eval_runtime": 12.6622, | |
| "eval_samples_per_second": 20.613, | |
| "eval_steps_per_second": 0.711, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 10.27, | |
| "learning_rate": 2.623456790123457e-05, | |
| "loss": 0.3381, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "learning_rate": 2.4691358024691357e-05, | |
| "loss": 0.289, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "eval_accuracy": 0.7739463601532567, | |
| "eval_loss": 0.5579754114151001, | |
| "eval_runtime": 13.0103, | |
| "eval_samples_per_second": 20.061, | |
| "eval_steps_per_second": 0.692, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.2695, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 11.89, | |
| "learning_rate": 2.1604938271604937e-05, | |
| "loss": 0.2827, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7969348659003831, | |
| "eval_loss": 0.5307646989822388, | |
| "eval_runtime": 12.5136, | |
| "eval_samples_per_second": 20.857, | |
| "eval_steps_per_second": 0.719, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 12.43, | |
| "learning_rate": 2.006172839506173e-05, | |
| "loss": 0.2516, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 12.97, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.2375, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 12.97, | |
| "eval_accuracy": 0.8045977011494253, | |
| "eval_loss": 0.5273889899253845, | |
| "eval_runtime": 12.9936, | |
| "eval_samples_per_second": 20.087, | |
| "eval_steps_per_second": 0.693, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 13.51, | |
| "learning_rate": 1.697530864197531e-05, | |
| "loss": 0.2493, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8045977011494253, | |
| "eval_loss": 0.5433180332183838, | |
| "eval_runtime": 12.7166, | |
| "eval_samples_per_second": 20.524, | |
| "eval_steps_per_second": 0.708, | |
| "step": 259 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "learning_rate": 1.54320987654321e-05, | |
| "loss": 0.2301, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 14.59, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.2309, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_accuracy": 0.7931034482758621, | |
| "eval_loss": 0.5354548692703247, | |
| "eval_runtime": 13.0668, | |
| "eval_samples_per_second": 19.974, | |
| "eval_steps_per_second": 0.689, | |
| "step": 277 | |
| }, | |
| { | |
| "epoch": 15.14, | |
| "learning_rate": 1.2345679012345678e-05, | |
| "loss": 0.2187, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 15.68, | |
| "learning_rate": 1.0802469135802469e-05, | |
| "loss": 0.1963, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8314176245210728, | |
| "eval_loss": 0.483632355928421, | |
| "eval_runtime": 13.0481, | |
| "eval_samples_per_second": 20.003, | |
| "eval_steps_per_second": 0.69, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 16.22, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.2209, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 16.76, | |
| "learning_rate": 7.71604938271605e-06, | |
| "loss": 0.2162, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 16.97, | |
| "eval_accuracy": 0.8237547892720306, | |
| "eval_loss": 0.4972710609436035, | |
| "eval_runtime": 13.0793, | |
| "eval_samples_per_second": 19.955, | |
| "eval_steps_per_second": 0.688, | |
| "step": 314 | |
| }, | |
| { | |
| "epoch": 17.3, | |
| "learning_rate": 6.172839506172839e-06, | |
| "loss": 0.1948, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 17.84, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.2256, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8275862068965517, | |
| "eval_loss": 0.49177083373069763, | |
| "eval_runtime": 13.1863, | |
| "eval_samples_per_second": 19.793, | |
| "eval_steps_per_second": 0.683, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 18.38, | |
| "learning_rate": 3.0864197530864196e-06, | |
| "loss": 0.2093, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 18.92, | |
| "learning_rate": 1.5432098765432098e-06, | |
| "loss": 0.2124, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "eval_accuracy": 0.8160919540229885, | |
| "eval_loss": 0.5070743560791016, | |
| "eval_runtime": 13.1883, | |
| "eval_samples_per_second": 19.79, | |
| "eval_steps_per_second": 0.682, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 19.46, | |
| "learning_rate": 0.0, | |
| "loss": 0.1797, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 19.46, | |
| "eval_accuracy": 0.8199233716475096, | |
| "eval_loss": 0.4984593987464905, | |
| "eval_runtime": 12.9999, | |
| "eval_samples_per_second": 20.077, | |
| "eval_steps_per_second": 0.692, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 19.46, | |
| "step": 360, | |
| "total_flos": 1.133589052022999e+18, | |
| "train_loss": 0.422045146756702, | |
| "train_runtime": 3802.191, | |
| "train_samples_per_second": 12.324, | |
| "train_steps_per_second": 0.095 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 360, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 1.133589052022999e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |