| { | |
| "best_metric": 0.75, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-dmae-va-U5-42\\checkpoint-108", | |
| "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": 0.45, | |
| "eval_loss": 1.383138656616211, | |
| "eval_runtime": 0.8504, | |
| "eval_samples_per_second": 70.553, | |
| "eval_steps_per_second": 2.352, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.3852, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.45, | |
| "eval_loss": 1.3624320030212402, | |
| "eval_runtime": 0.8336, | |
| "eval_samples_per_second": 71.981, | |
| "eval_steps_per_second": 2.399, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.3728, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.4666666666666667, | |
| "eval_loss": 1.2926621437072754, | |
| "eval_runtime": 0.8819, | |
| "eval_samples_per_second": 68.033, | |
| "eval_steps_per_second": 2.268, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2791, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.48333333333333334, | |
| "eval_loss": 1.1181586980819702, | |
| "eval_runtime": 0.9023, | |
| "eval_samples_per_second": 66.499, | |
| "eval_steps_per_second": 2.217, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_accuracy": 0.4666666666666667, | |
| "eval_loss": 1.0064685344696045, | |
| "eval_runtime": 1.1345, | |
| "eval_samples_per_second": 52.886, | |
| "eval_steps_per_second": 1.763, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 4.810606060606061e-05, | |
| "loss": 1.094, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 0.8931151032447815, | |
| "eval_runtime": 0.9014, | |
| "eval_samples_per_second": 66.566, | |
| "eval_steps_per_second": 2.219, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 4.621212121212121e-05, | |
| "loss": 0.9601, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.48333333333333334, | |
| "eval_loss": 0.9209610819816589, | |
| "eval_runtime": 1.4464, | |
| "eval_samples_per_second": 41.482, | |
| "eval_steps_per_second": 1.383, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 4.431818181818182e-05, | |
| "loss": 0.8598, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5166666666666667, | |
| "eval_loss": 0.947771430015564, | |
| "eval_runtime": 1.4436, | |
| "eval_samples_per_second": 41.563, | |
| "eval_steps_per_second": 1.385, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.8557999134063721, | |
| "eval_runtime": 1.0161, | |
| "eval_samples_per_second": 59.05, | |
| "eval_steps_per_second": 1.968, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 4.242424242424243e-05, | |
| "loss": 0.7558, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "eval_accuracy": 0.65, | |
| "eval_loss": 0.9258978366851807, | |
| "eval_runtime": 0.9985, | |
| "eval_samples_per_second": 60.091, | |
| "eval_steps_per_second": 2.003, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 4.053030303030303e-05, | |
| "loss": 0.6696, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "eval_accuracy": 0.6166666666666667, | |
| "eval_loss": 0.7952563762664795, | |
| "eval_runtime": 1.3622, | |
| "eval_samples_per_second": 44.048, | |
| "eval_steps_per_second": 1.468, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 3.8636363636363636e-05, | |
| "loss": 0.6079, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7, | |
| "eval_loss": 0.7281058430671692, | |
| "eval_runtime": 1.1664, | |
| "eval_samples_per_second": 51.442, | |
| "eval_steps_per_second": 1.715, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 3.6742424242424246e-05, | |
| "loss": 0.516, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 0.8551703691482544, | |
| "eval_runtime": 2.0248, | |
| "eval_samples_per_second": 29.633, | |
| "eval_steps_per_second": 0.988, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.6653277277946472, | |
| "eval_runtime": 3.2967, | |
| "eval_samples_per_second": 18.2, | |
| "eval_steps_per_second": 0.607, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 3.484848484848485e-05, | |
| "loss": 0.4475, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_accuracy": 0.6833333333333333, | |
| "eval_loss": 0.7548192739486694, | |
| "eval_runtime": 1.3318, | |
| "eval_samples_per_second": 45.053, | |
| "eval_steps_per_second": 1.502, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 3.295454545454545e-05, | |
| "loss": 0.4152, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7166666666666667, | |
| "eval_loss": 0.7556443810462952, | |
| "eval_runtime": 2.1003, | |
| "eval_samples_per_second": 28.568, | |
| "eval_steps_per_second": 0.952, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 16.77, | |
| "learning_rate": 3.106060606060606e-05, | |
| "loss": 0.3759, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "eval_accuracy": 0.7333333333333333, | |
| "eval_loss": 0.7037982940673828, | |
| "eval_runtime": 1.1281, | |
| "eval_samples_per_second": 53.188, | |
| "eval_steps_per_second": 1.773, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "eval_accuracy": 0.7166666666666667, | |
| "eval_loss": 0.7355785369873047, | |
| "eval_runtime": 1.7841, | |
| "eval_samples_per_second": 33.631, | |
| "eval_steps_per_second": 1.121, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 0.3366, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.6649565100669861, | |
| "eval_runtime": 1.014, | |
| "eval_samples_per_second": 59.17, | |
| "eval_steps_per_second": 1.972, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 0.3212, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7166666666666667, | |
| "eval_loss": 0.7667668461799622, | |
| "eval_runtime": 1.2664, | |
| "eval_samples_per_second": 47.379, | |
| "eval_steps_per_second": 1.579, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 20.65, | |
| "learning_rate": 2.537878787878788e-05, | |
| "loss": 0.2903, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "eval_accuracy": 0.6833333333333333, | |
| "eval_loss": 0.7996882796287537, | |
| "eval_runtime": 0.8672, | |
| "eval_samples_per_second": 69.189, | |
| "eval_steps_per_second": 2.306, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "learning_rate": 2.3484848484848487e-05, | |
| "loss": 0.312, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "eval_accuracy": 0.7166666666666667, | |
| "eval_loss": 0.7473248243331909, | |
| "eval_runtime": 0.9023, | |
| "eval_samples_per_second": 66.497, | |
| "eval_steps_per_second": 2.217, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "eval_accuracy": 0.7, | |
| "eval_loss": 0.8479262590408325, | |
| "eval_runtime": 0.7981, | |
| "eval_samples_per_second": 75.176, | |
| "eval_steps_per_second": 2.506, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 2.1590909090909093e-05, | |
| "loss": 0.2488, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7, | |
| "eval_loss": 0.8069732785224915, | |
| "eval_runtime": 1.0488, | |
| "eval_samples_per_second": 57.211, | |
| "eval_steps_per_second": 1.907, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 24.52, | |
| "learning_rate": 1.9696969696969697e-05, | |
| "loss": 0.283, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 24.9, | |
| "eval_accuracy": 0.6833333333333333, | |
| "eval_loss": 0.8079617619514465, | |
| "eval_runtime": 1.2667, | |
| "eval_samples_per_second": 47.366, | |
| "eval_steps_per_second": 1.579, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "learning_rate": 1.7803030303030303e-05, | |
| "loss": 0.2109, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 25.94, | |
| "eval_accuracy": 0.7, | |
| "eval_loss": 0.8220053315162659, | |
| "eval_runtime": 1.2171, | |
| "eval_samples_per_second": 49.296, | |
| "eval_steps_per_second": 1.643, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 26.97, | |
| "eval_accuracy": 0.7166666666666667, | |
| "eval_loss": 0.8354397416114807, | |
| "eval_runtime": 0.8636, | |
| "eval_samples_per_second": 69.476, | |
| "eval_steps_per_second": 2.316, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 27.1, | |
| "learning_rate": 1.590909090909091e-05, | |
| "loss": 0.2215, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.7, | |
| "eval_loss": 0.8667593598365784, | |
| "eval_runtime": 1.1633, | |
| "eval_samples_per_second": 51.577, | |
| "eval_steps_per_second": 1.719, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 28.39, | |
| "learning_rate": 1.4015151515151515e-05, | |
| "loss": 0.2067, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 28.9, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.8478964567184448, | |
| "eval_runtime": 1.6328, | |
| "eval_samples_per_second": 36.746, | |
| "eval_steps_per_second": 1.225, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 29.68, | |
| "learning_rate": 1.2121212121212122e-05, | |
| "loss": 0.1967, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 29.94, | |
| "eval_accuracy": 0.7166666666666667, | |
| "eval_loss": 0.8867913484573364, | |
| "eval_runtime": 1.015, | |
| "eval_samples_per_second": 59.114, | |
| "eval_steps_per_second": 1.97, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "learning_rate": 1.0227272727272729e-05, | |
| "loss": 0.1948, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "eval_accuracy": 0.7, | |
| "eval_loss": 0.8882503509521484, | |
| "eval_runtime": 1.8662, | |
| "eval_samples_per_second": 32.152, | |
| "eval_steps_per_second": 1.072, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7333333333333333, | |
| "eval_loss": 0.8612008690834045, | |
| "eval_runtime": 1.182, | |
| "eval_samples_per_second": 50.764, | |
| "eval_steps_per_second": 1.692, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.186, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 32.9, | |
| "eval_accuracy": 0.7333333333333333, | |
| "eval_loss": 0.8859888315200806, | |
| "eval_runtime": 1.7167, | |
| "eval_samples_per_second": 34.951, | |
| "eval_steps_per_second": 1.165, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 33.55, | |
| "learning_rate": 6.43939393939394e-06, | |
| "loss": 0.1662, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 33.94, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.9057492613792419, | |
| "eval_runtime": 0.9173, | |
| "eval_samples_per_second": 65.409, | |
| "eval_steps_per_second": 2.18, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 34.84, | |
| "learning_rate": 4.5454545454545455e-06, | |
| "loss": 0.1773, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 34.97, | |
| "eval_accuracy": 0.7166666666666667, | |
| "eval_loss": 0.9140269160270691, | |
| "eval_runtime": 1.1633, | |
| "eval_samples_per_second": 51.579, | |
| "eval_steps_per_second": 1.719, | |
| "step": 271 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.7333333333333333, | |
| "eval_loss": 0.9013189673423767, | |
| "eval_runtime": 0.8518, | |
| "eval_samples_per_second": 70.443, | |
| "eval_steps_per_second": 2.348, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "learning_rate": 2.651515151515152e-06, | |
| "loss": 0.1519, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 36.9, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.8868784308433533, | |
| "eval_runtime": 0.8675, | |
| "eval_samples_per_second": 69.164, | |
| "eval_steps_per_second": 2.305, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 37.42, | |
| "learning_rate": 7.575757575757576e-07, | |
| "loss": 0.1775, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 37.94, | |
| "eval_accuracy": 0.7333333333333333, | |
| "eval_loss": 0.8840016722679138, | |
| "eval_runtime": 1.1992, | |
| "eval_samples_per_second": 50.034, | |
| "eval_steps_per_second": 1.668, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 37.94, | |
| "step": 294, | |
| "total_flos": 9.188778373008998e+17, | |
| "train_loss": 0.4929502831429851, | |
| "train_runtime": 549.9582, | |
| "train_samples_per_second": 74.384, | |
| "train_steps_per_second": 0.535 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 294, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 42, | |
| "save_steps": 500, | |
| "total_flos": 9.188778373008998e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |