| { | |
| "best_metric": 0.65, | |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-va-U5-42B\\checkpoint-209", | |
| "epoch": 37.935483870967744, | |
| "eval_steps": 500, | |
| "global_step": 294, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 6.174824237823486, | |
| "eval_runtime": 0.845, | |
| "eval_samples_per_second": 71.01, | |
| "eval_steps_per_second": 2.367, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 5.327, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 6.066026210784912, | |
| "eval_runtime": 0.8511, | |
| "eval_samples_per_second": 70.499, | |
| "eval_steps_per_second": 2.35, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 5.4902424812316895, | |
| "eval_runtime": 0.8648, | |
| "eval_samples_per_second": 69.381, | |
| "eval_steps_per_second": 2.313, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 5.0963, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 4.276790142059326, | |
| "eval_runtime": 0.8095, | |
| "eval_samples_per_second": 74.116, | |
| "eval_steps_per_second": 2.471, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 3.909090909090909e-05, | |
| "loss": 3.9193, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 3.0012993812561035, | |
| "eval_runtime": 0.8424, | |
| "eval_samples_per_second": 71.228, | |
| "eval_steps_per_second": 2.374, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 1.9289205074310303, | |
| "eval_runtime": 0.8816, | |
| "eval_samples_per_second": 68.055, | |
| "eval_steps_per_second": 2.269, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 3.7272727272727276e-05, | |
| "loss": 2.2222, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 1.385749101638794, | |
| "eval_runtime": 0.8792, | |
| "eval_samples_per_second": 68.244, | |
| "eval_steps_per_second": 2.275, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 3.545454545454546e-05, | |
| "loss": 1.4465, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.43333333333333335, | |
| "eval_loss": 1.342329740524292, | |
| "eval_runtime": 0.8314, | |
| "eval_samples_per_second": 72.165, | |
| "eval_steps_per_second": 2.405, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "eval_accuracy": 0.45, | |
| "eval_loss": 1.2786400318145752, | |
| "eval_runtime": 0.8842, | |
| "eval_samples_per_second": 67.857, | |
| "eval_steps_per_second": 2.262, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 9.29, | |
| "learning_rate": 3.363636363636364e-05, | |
| "loss": 1.3709, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "eval_accuracy": 0.4666666666666667, | |
| "eval_loss": 1.2653944492340088, | |
| "eval_runtime": 0.8365, | |
| "eval_samples_per_second": 71.724, | |
| "eval_steps_per_second": 2.391, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 10.84, | |
| "learning_rate": 3.181818181818182e-05, | |
| "loss": 1.3511, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "eval_accuracy": 0.4666666666666667, | |
| "eval_loss": 1.2605175971984863, | |
| "eval_runtime": 0.8152, | |
| "eval_samples_per_second": 73.599, | |
| "eval_steps_per_second": 2.453, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4666666666666667, | |
| "eval_loss": 1.2183691263198853, | |
| "eval_runtime": 0.8329, | |
| "eval_samples_per_second": 72.04, | |
| "eval_steps_per_second": 2.401, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 12.39, | |
| "learning_rate": 3.0000000000000004e-05, | |
| "loss": 1.2749, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.289415955543518, | |
| "eval_runtime": 0.8348, | |
| "eval_samples_per_second": 71.87, | |
| "eval_steps_per_second": 2.396, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "learning_rate": 2.8181818181818185e-05, | |
| "loss": 1.222, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "eval_accuracy": 0.5166666666666667, | |
| "eval_loss": 1.2071518898010254, | |
| "eval_runtime": 0.8805, | |
| "eval_samples_per_second": 68.142, | |
| "eval_steps_per_second": 2.271, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_accuracy": 0.5166666666666667, | |
| "eval_loss": 1.1748946905136108, | |
| "eval_runtime": 0.8381, | |
| "eval_samples_per_second": 71.592, | |
| "eval_steps_per_second": 2.386, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 2.6363636363636365e-05, | |
| "loss": 1.1668, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5166666666666667, | |
| "eval_loss": 1.198820948600769, | |
| "eval_runtime": 0.8663, | |
| "eval_samples_per_second": 69.26, | |
| "eval_steps_per_second": 2.309, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "eval_accuracy": 0.5166666666666667, | |
| "eval_loss": 1.2305818796157837, | |
| "eval_runtime": 0.8839, | |
| "eval_samples_per_second": 67.884, | |
| "eval_steps_per_second": 2.263, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 17.03, | |
| "learning_rate": 2.454545454545455e-05, | |
| "loss": 1.101, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "eval_accuracy": 0.5333333333333333, | |
| "eval_loss": 1.1431663036346436, | |
| "eval_runtime": 0.8778, | |
| "eval_samples_per_second": 68.351, | |
| "eval_steps_per_second": 2.278, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 18.58, | |
| "learning_rate": 2.2727272727272733e-05, | |
| "loss": 1.029, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "eval_accuracy": 0.55, | |
| "eval_loss": 1.0208261013031006, | |
| "eval_runtime": 0.8723, | |
| "eval_samples_per_second": 68.786, | |
| "eval_steps_per_second": 2.293, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.6166666666666667, | |
| "eval_loss": 0.9576696157455444, | |
| "eval_runtime": 0.849, | |
| "eval_samples_per_second": 70.671, | |
| "eval_steps_per_second": 2.356, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 20.13, | |
| "learning_rate": 2.090909090909091e-05, | |
| "loss": 0.9403, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.9478870034217834, | |
| "eval_runtime": 0.8337, | |
| "eval_samples_per_second": 71.971, | |
| "eval_steps_per_second": 2.399, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 21.68, | |
| "learning_rate": 1.9090909090909094e-05, | |
| "loss": 0.8887, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.8909610509872437, | |
| "eval_runtime": 0.8389, | |
| "eval_samples_per_second": 71.521, | |
| "eval_steps_per_second": 2.384, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "eval_accuracy": 0.5333333333333333, | |
| "eval_loss": 0.9441585540771484, | |
| "eval_runtime": 0.8343, | |
| "eval_samples_per_second": 71.914, | |
| "eval_steps_per_second": 2.397, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 1.7272727272727274e-05, | |
| "loss": 0.8506, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 0.8923367261886597, | |
| "eval_runtime": 0.8341, | |
| "eval_samples_per_second": 71.937, | |
| "eval_steps_per_second": 2.398, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 24.77, | |
| "learning_rate": 1.5454545454545454e-05, | |
| "loss": 0.8064, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 24.9, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 0.8972867131233215, | |
| "eval_runtime": 0.8817, | |
| "eval_samples_per_second": 68.052, | |
| "eval_steps_per_second": 2.268, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 25.94, | |
| "eval_accuracy": 0.55, | |
| "eval_loss": 0.907895028591156, | |
| "eval_runtime": 0.87, | |
| "eval_samples_per_second": 68.962, | |
| "eval_steps_per_second": 2.299, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "learning_rate": 1.3636363636363637e-05, | |
| "loss": 0.7434, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 26.97, | |
| "eval_accuracy": 0.65, | |
| "eval_loss": 0.8386151194572449, | |
| "eval_runtime": 0.9123, | |
| "eval_samples_per_second": 65.768, | |
| "eval_steps_per_second": 2.192, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 27.87, | |
| "learning_rate": 1.181818181818182e-05, | |
| "loss": 0.7404, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.6166666666666667, | |
| "eval_loss": 0.8644894361495972, | |
| "eval_runtime": 0.8305, | |
| "eval_samples_per_second": 72.242, | |
| "eval_steps_per_second": 2.408, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 28.9, | |
| "eval_accuracy": 0.5666666666666667, | |
| "eval_loss": 0.8599310517311096, | |
| "eval_runtime": 0.9003, | |
| "eval_samples_per_second": 66.644, | |
| "eval_steps_per_second": 2.221, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 29.42, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7215, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 29.94, | |
| "eval_accuracy": 0.65, | |
| "eval_loss": 0.8420030474662781, | |
| "eval_runtime": 0.8701, | |
| "eval_samples_per_second": 68.954, | |
| "eval_steps_per_second": 2.298, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "learning_rate": 8.181818181818183e-06, | |
| "loss": 0.6743, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "eval_accuracy": 0.5666666666666667, | |
| "eval_loss": 0.855254590511322, | |
| "eval_runtime": 0.8716, | |
| "eval_samples_per_second": 68.841, | |
| "eval_steps_per_second": 2.295, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.6166666666666667, | |
| "eval_loss": 0.8355186581611633, | |
| "eval_runtime": 0.8931, | |
| "eval_samples_per_second": 67.182, | |
| "eval_steps_per_second": 2.239, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 32.52, | |
| "learning_rate": 6.363636363636364e-06, | |
| "loss": 0.6767, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 32.9, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.8694211840629578, | |
| "eval_runtime": 0.87, | |
| "eval_samples_per_second": 68.966, | |
| "eval_steps_per_second": 2.299, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 33.94, | |
| "eval_accuracy": 0.65, | |
| "eval_loss": 0.8559350371360779, | |
| "eval_runtime": 0.8505, | |
| "eval_samples_per_second": 70.545, | |
| "eval_steps_per_second": 2.351, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 34.06, | |
| "learning_rate": 4.5454545454545455e-06, | |
| "loss": 0.6606, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 34.97, | |
| "eval_accuracy": 0.6166666666666667, | |
| "eval_loss": 0.8350656032562256, | |
| "eval_runtime": 0.8389, | |
| "eval_samples_per_second": 71.519, | |
| "eval_steps_per_second": 2.384, | |
| "step": 271 | |
| }, | |
| { | |
| "epoch": 35.61, | |
| "learning_rate": 2.7272727272727272e-06, | |
| "loss": 0.6488, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.6333333333333333, | |
| "eval_loss": 0.8287425637245178, | |
| "eval_runtime": 0.8966, | |
| "eval_samples_per_second": 66.919, | |
| "eval_steps_per_second": 2.231, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 36.9, | |
| "eval_accuracy": 0.6166666666666667, | |
| "eval_loss": 0.8377331495285034, | |
| "eval_runtime": 0.8452, | |
| "eval_samples_per_second": 70.987, | |
| "eval_steps_per_second": 2.366, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 37.16, | |
| "learning_rate": 9.090909090909091e-07, | |
| "loss": 0.6544, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 37.94, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 0.8406462073326111, | |
| "eval_runtime": 0.8978, | |
| "eval_samples_per_second": 66.829, | |
| "eval_steps_per_second": 2.228, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 37.94, | |
| "step": 294, | |
| "total_flos": 1.2027310550050406e+18, | |
| "train_loss": 1.4638761638784084, | |
| "train_runtime": 475.2898, | |
| "train_samples_per_second": 86.07, | |
| "train_steps_per_second": 0.619 | |
| } | |
| ], | |
| "logging_steps": 12, | |
| "max_steps": 294, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 42, | |
| "save_steps": 500, | |
| "total_flos": 1.2027310550050406e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |