| { | |
| "best_metric": 0.6822429906542056, | |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-RH\\checkpoint-296", | |
| "epoch": 40.0, | |
| "eval_steps": 500, | |
| "global_step": 320, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.411214953271028, | |
| "eval_loss": 4.626469612121582, | |
| "eval_runtime": 1.941, | |
| "eval_samples_per_second": 55.127, | |
| "eval_steps_per_second": 3.606, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1.25e-05, | |
| "loss": 4.5369, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.411214953271028, | |
| "eval_loss": 4.52950382232666, | |
| "eval_runtime": 1.8274, | |
| "eval_samples_per_second": 58.552, | |
| "eval_steps_per_second": 3.831, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 2.5e-05, | |
| "loss": 4.6305, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.411214953271028, | |
| "eval_loss": 4.143911361694336, | |
| "eval_runtime": 1.8444, | |
| "eval_samples_per_second": 58.012, | |
| "eval_steps_per_second": 3.795, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 4.0918, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.411214953271028, | |
| "eval_loss": 3.369269609451294, | |
| "eval_runtime": 1.8599, | |
| "eval_samples_per_second": 57.529, | |
| "eval_steps_per_second": 3.764, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 3.1767, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.411214953271028, | |
| "eval_loss": 2.4324560165405273, | |
| "eval_runtime": 1.894, | |
| "eval_samples_per_second": 56.496, | |
| "eval_steps_per_second": 3.696, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.411214953271028, | |
| "eval_loss": 1.5421515703201294, | |
| "eval_runtime": 1.8605, | |
| "eval_samples_per_second": 57.513, | |
| "eval_steps_per_second": 3.763, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 2.0113, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.411214953271028, | |
| "eval_loss": 0.8834472894668579, | |
| "eval_runtime": 1.8524, | |
| "eval_samples_per_second": 57.762, | |
| "eval_steps_per_second": 3.779, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 3.6111111111111116e-05, | |
| "loss": 1.0593, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.6790158748626709, | |
| "eval_runtime": 1.8709, | |
| "eval_samples_per_second": 57.19, | |
| "eval_steps_per_second": 3.741, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "learning_rate": 3.472222222222223e-05, | |
| "loss": 0.696, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.7044364213943481, | |
| "eval_runtime": 1.8454, | |
| "eval_samples_per_second": 57.981, | |
| "eval_steps_per_second": 3.793, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.6893, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.6777833104133606, | |
| "eval_runtime": 1.8339, | |
| "eval_samples_per_second": 58.344, | |
| "eval_steps_per_second": 3.817, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.686574399471283, | |
| "eval_runtime": 1.8689, | |
| "eval_samples_per_second": 57.252, | |
| "eval_steps_per_second": 3.745, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "learning_rate": 3.194444444444445e-05, | |
| "loss": 0.6961, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.6933708786964417, | |
| "eval_runtime": 1.8659, | |
| "eval_samples_per_second": 57.344, | |
| "eval_steps_per_second": 3.751, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 3.0555555555555554e-05, | |
| "loss": 0.7329, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.6915177702903748, | |
| "eval_runtime": 1.8689, | |
| "eval_samples_per_second": 57.252, | |
| "eval_steps_per_second": 3.745, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "learning_rate": 2.9166666666666666e-05, | |
| "loss": 0.6948, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.6761561036109924, | |
| "eval_runtime": 1.873, | |
| "eval_samples_per_second": 57.129, | |
| "eval_steps_per_second": 3.737, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.6771, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.6794873476028442, | |
| "eval_runtime": 1.8549, | |
| "eval_samples_per_second": 57.684, | |
| "eval_steps_per_second": 3.774, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.6800874471664429, | |
| "eval_runtime": 1.871, | |
| "eval_samples_per_second": 57.19, | |
| "eval_steps_per_second": 3.741, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "learning_rate": 2.6388888888888892e-05, | |
| "loss": 0.6763, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.6819825172424316, | |
| "eval_runtime": 1.8614, | |
| "eval_samples_per_second": 57.483, | |
| "eval_steps_per_second": 3.761, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6822, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.6799528002738953, | |
| "eval_runtime": 1.8664, | |
| "eval_samples_per_second": 57.328, | |
| "eval_steps_per_second": 3.75, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "learning_rate": 2.3611111111111114e-05, | |
| "loss": 0.6723, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.6740810871124268, | |
| "eval_runtime": 1.8689, | |
| "eval_samples_per_second": 57.252, | |
| "eval_steps_per_second": 3.745, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 2.2222222222222227e-05, | |
| "loss": 0.6757, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.681536078453064, | |
| "eval_runtime": 1.8439, | |
| "eval_samples_per_second": 58.028, | |
| "eval_steps_per_second": 3.796, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.6728676557540894, | |
| "eval_runtime": 1.8774, | |
| "eval_samples_per_second": 56.993, | |
| "eval_steps_per_second": 3.729, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.6711, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.6811678409576416, | |
| "eval_runtime": 1.8474, | |
| "eval_samples_per_second": 57.918, | |
| "eval_steps_per_second": 3.789, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 22.5, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.6784, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.5794392523364486, | |
| "eval_loss": 0.6780757308006287, | |
| "eval_runtime": 1.9144, | |
| "eval_samples_per_second": 55.891, | |
| "eval_steps_per_second": 3.656, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 23.75, | |
| "learning_rate": 1.8055555555555558e-05, | |
| "loss": 0.6665, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.5794392523364486, | |
| "eval_loss": 0.6697570085525513, | |
| "eval_runtime": 1.8669, | |
| "eval_samples_per_second": 57.313, | |
| "eval_steps_per_second": 3.749, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.6723, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.5981308411214953, | |
| "eval_loss": 0.6646633744239807, | |
| "eval_runtime": 1.8769, | |
| "eval_samples_per_second": 57.008, | |
| "eval_steps_per_second": 3.729, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.5794392523364486, | |
| "eval_loss": 0.6761506199836731, | |
| "eval_runtime": 1.8489, | |
| "eval_samples_per_second": 57.871, | |
| "eval_steps_per_second": 3.786, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 26.25, | |
| "learning_rate": 1.5277777777777777e-05, | |
| "loss": 0.6675, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.5700934579439252, | |
| "eval_loss": 0.659662127494812, | |
| "eval_runtime": 1.8699, | |
| "eval_samples_per_second": 57.221, | |
| "eval_steps_per_second": 3.743, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 27.5, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.6628, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.6355140186915887, | |
| "eval_loss": 0.6563044190406799, | |
| "eval_runtime": 1.901, | |
| "eval_samples_per_second": 56.288, | |
| "eval_steps_per_second": 3.682, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 28.75, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.6478, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.5794392523364486, | |
| "eval_loss": 0.6790612936019897, | |
| "eval_runtime": 1.8634, | |
| "eval_samples_per_second": 57.421, | |
| "eval_steps_per_second": 3.757, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 1.1111111111111113e-05, | |
| "loss": 0.6642, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.6573609709739685, | |
| "eval_runtime": 1.884, | |
| "eval_samples_per_second": 56.796, | |
| "eval_steps_per_second": 3.716, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.5607476635514018, | |
| "eval_loss": 0.6556074023246765, | |
| "eval_runtime": 1.872, | |
| "eval_samples_per_second": 57.159, | |
| "eval_steps_per_second": 3.739, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 31.25, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.654, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.5887850467289719, | |
| "eval_loss": 0.6523457169532776, | |
| "eval_runtime": 1.871, | |
| "eval_samples_per_second": 57.19, | |
| "eval_steps_per_second": 3.741, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 32.5, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.6602, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.6261682242990654, | |
| "eval_loss": 0.6463862657546997, | |
| "eval_runtime": 1.8604, | |
| "eval_samples_per_second": 57.513, | |
| "eval_steps_per_second": 3.763, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 33.75, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.6535, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.616822429906542, | |
| "eval_loss": 0.6450387835502625, | |
| "eval_runtime": 2.0415, | |
| "eval_samples_per_second": 52.413, | |
| "eval_steps_per_second": 3.429, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "learning_rate": 5.555555555555557e-06, | |
| "loss": 0.6506, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.5794392523364486, | |
| "eval_loss": 0.6549742221832275, | |
| "eval_runtime": 1.8524, | |
| "eval_samples_per_second": 57.762, | |
| "eval_steps_per_second": 3.779, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.6074766355140186, | |
| "eval_loss": 0.6437696218490601, | |
| "eval_runtime": 1.837, | |
| "eval_samples_per_second": 58.249, | |
| "eval_steps_per_second": 3.811, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 36.25, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.6533, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.6822429906542056, | |
| "eval_loss": 0.6395505666732788, | |
| "eval_runtime": 1.8519, | |
| "eval_samples_per_second": 57.777, | |
| "eval_steps_per_second": 3.78, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 37.5, | |
| "learning_rate": 2.7777777777777783e-06, | |
| "loss": 0.6443, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.6635514018691588, | |
| "eval_loss": 0.6383424997329712, | |
| "eval_runtime": 1.8994, | |
| "eval_samples_per_second": 56.332, | |
| "eval_steps_per_second": 3.685, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 38.75, | |
| "learning_rate": 1.3888888888888892e-06, | |
| "loss": 0.6263, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.6448598130841121, | |
| "eval_loss": 0.6377760171890259, | |
| "eval_runtime": 1.8969, | |
| "eval_samples_per_second": 56.406, | |
| "eval_steps_per_second": 3.69, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.6283, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.6448598130841121, | |
| "eval_loss": 0.6378776431083679, | |
| "eval_runtime": 1.8389, | |
| "eval_samples_per_second": 58.186, | |
| "eval_steps_per_second": 3.807, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "step": 320, | |
| "total_flos": 6.480838238876467e+17, | |
| "train_loss": 1.1531404286623002, | |
| "train_runtime": 569.0029, | |
| "train_samples_per_second": 35.009, | |
| "train_steps_per_second": 0.562 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 320, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 6.480838238876467e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |